Advanced Global Wildfire Activity Modeling with Hierarchical Graph ODE

TL;DR

The paper reframes global wildfire forecasting as a multi-scale, continuous-time problem and introduces HiGO, a Hierarchical Graph ODE framework that architectures a multi-level graph of the Earth system, uses climate-informed gated fusion to fuse driver variables with global indices, and evolves the state via a neural ODE with adaptive intra- and inter-level message passing. The model integrates a hierarchical graph ODE within the HiGO framework and a cross-attention-based feature mixer to produce ordinal, grid-level burn maps, trained with a weighted cross-entropy objective on SeasFire Cube data. Empirical evaluation demonstrates state-of-the-art long-range accuracy (up to 48 days) and strong continuous-time flexibility, outperforming vision- and graph-based baselines, with ablations highlighting the importance of land-driven features, adaptive message passing, and optimal hierarchy depth. The work advances practical, physically-consistent global wildfire forecasting with scalable, multi-scale dynamics that can adapt to irregular time steps and diverse forcing factors, enabling more reliable risk assessment and decision support.

Abstract

Wildfires, as an integral component of the Earth system, are governed by a complex interplay of atmospheric, oceanic, and terrestrial processes spanning a vast range of spatiotemporal scales. Modeling their global activity on large timescales is therefore a critical yet challenging task. While deep learning has recently achieved significant breakthroughs in global weather forecasting, its potential for global wildfire behavior prediction remains underexplored. In this work, we reframe this problem and introduce the Hierarchical Graph ODE (HiGO), a novel framework designed to learn the multi-scale, continuous-time dynamics of wildfires. Specifically, we represent the Earth system as a multi-level graph hierarchy and propose an adaptive filtering message passing mechanism for both intra- and inter-level information flow, enabling more effective feature extraction and fusion. Furthermore, we incorporate GNN-parameterized Neural ODE modules at multiple levels to explicitly learn the continuous dynamics inherent to each scale. Through extensive experiments on the SeasFire Cube dataset, we demonstrate that HiGO significantly outperforms state-of-the-art baselines on long-range wildfire forecasting. Moreover, its continuous-time predictions exhibit strong observational consistency, highlighting its potential for real-world applications.

Figures (5)

• Figure 1: The left depicts a local wildfire event, influenced by immediate conditions like airstreams. The right illustrates the broader Earth system context, framing wildfires as a phenomenon governed by a complex interaction between the Atmosphere, Ocean, Land and Human Activity.
• Figure 2: Architectural overview of the HiGO framework. The framework integrates multi-modal driver and climate data to initialize a hierarchical graph representation. The core of the model consists of a multi-scale graph pyramid, where HGODE blocks learn continuous-time dynamics and an adaptive message-passing scheme captures complex spatial dependencies.
• Figure 3: Visualization evaluation of HiGO's long-term wildfire activity forecasting performance. The panel provides a visual comparison between HiGO's multi-step predictions (top row) and the ground truth (bottom row) for up to a 48-day horizon, demonstrating high qualitative fidelity.
• Figure 4: Impact of Forcing Variables on 24-day Forecast Performance. The bar chart shows the AUPRC score for the full model (left) and five variants, each with one group of input variables removed.
• Figure 5: Impact of the number of hierarchical graph levels ($L$) on forecasting performance.