Search

More frequent and widespread large fires are occurring in the western United States (US), yet reliable methods for predicting these fires, particularly with extended lead times and a high spatial resolution, remain challenging. In this study, we proposed an interpretable and accurate hybrid machine learning (ML) model, that explicitly represented the controls of fuel flammability, fuel availability, and human suppression effects on fires. The model demonstrated notable accuracy with a F1‐score of 0.846 ± 0.012, surpassing process‐driven fire danger indices and four commonly used ML models by up to 40% and 9%, respectively. More importantly, the ML model showed remarkably higher interpretability relative to other ML models. Specifically, by demystifying the "black box" of each ML model using the explainable AI techniques, we identified substantial structural differences across ML fire models, even among those with similar accuracy. The relationships between fires and their drivers, identified by our model, were aligned closer with established fire physical principles. The ML structural discrepancy led to diverse fire predictions and our model predictions exhibited greater consistency with actual fire occurrence. With the highly interpretable and accurate model, we revealed the strong compound effects from multiple climate variables related to evaporative demand, energy release component, temperature, and wind speed, on the dynamics of large fires and megafires in the western US. Our findings highlight the importance of assessing the structural integrity of models in addition to their accuracy. They also underscore the critical need to address the rise in compound climate extremes linked to large wildfires. Plain Language Summary: Anthropogenic climate change has caused more frequent and widespread large fires in the western United States (US). Interpretable and accurate methods that can predict large fires in advance and at a high spatial resolution are essential for more targeted fire risk mitigation strategies, but the trade‐off between accuracy and interpretability persists in current fire models. For example, machine learning (ML) fire models can be more accurate than process models but operate as "black‐boxes" with a poor understanding of fire physics. Process fire models are more physically interpretable but remain less accurate than ML models. This research offers a novel perspective on the tradeoff and quantification of model physical interpretability and accuracy for wildfire predictions in the western US. More importantly, this study proposes a more interpretable and accurate ML fire model that explicitly incorporates fire knowledge while leveraging data‐informed modeling advantages. The model predicted fire risk showed high consistency with actually occurred fire events, and revealed strong and multivariate climate controls on large fires and megafires in the western US. Our proposed fire model enables better predictions and understanding of large fires and megafires in the western US. Key Points: Substantial structural differences were identified across machine learning (ML) fire models, but existing studies mainly focus on accuracyWe proposed a fire model with higher accuracy and physical interpretability compared to commonly used ML modelsOur fire model reasonably reflected actual fire risk and revealed complex climate controls on large fires and megafires in the western US [ABSTRACT FROM AUTHOR]

Summary: Leaf traits are essential for understanding many physiological and ecological processes. Partial least squares regression (PLSR) models with leaf spectroscopy are widely applied for trait estimation, but their transferability across space, time, and plant functional types (PFTs) remains unclear.We compiled a novel dataset of paired leaf traits and spectra, with 47 393 records for > 700 species and eight PFTs at 101 globally distributed locations across multiple seasons. Using this dataset, we conducted an unprecedented comprehensive analysis to assess the transferability of PLSR models in estimating leaf traits.While PLSR models demonstrate commendable performance in predicting chlorophyll content, carotenoid, leaf water, and leaf mass per area prediction within their training data space, their efficacy diminishes when extrapolating to new contexts. Specifically, extrapolating to locations, seasons, and PFTs beyond the training data leads to reduced R2 (0.12–0.49, 0.15–0.42, and 0.25–0.56) and increased NRMSE (3.58–18.24%, 6.27–11.55%, and 7.0–33.12%) compared with nonspatial random cross‐validation. The results underscore the importance of incorporating greater spectral diversity in model training to boost its transferability.These findings highlight potential errors in estimating leaf traits across large spatial domains, diverse PFTs, and time due to biased validation schemes, and provide guidance for future field sampling strategies and remote sensing applications. [ABSTRACT FROM AUTHOR]