Search

African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire-climate relationship remains challenging, due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable Machine Learning (ML) fire model (AttentionFire_v1.0) to resolve the complex spatial- heterogenous and time-lagged controls from climate on burned area and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned area for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that under a high emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides reliable and interpretable fire model and highlights the importance of lagged wildfire-climate relationships in historical and future predictions.

Land surfaces dissipate energy through latent (LE) and sensible (H) heat fluxes that modulate atmospheric temperature and humidity, which in return affect land surface vegetation and soil processes. Within this two-way land-atmosphere coupling, surface energy partitioning (LE versus H) plays a central role in connecting the land and atmosphere states and fluxes. However, considerably large uncertainties still exist in earth system land models, i.e. the phase 6 of the Coupled Model Intercomparison Project (CMIP6). Further, the underlying controls from climate and biological factors on surface energy partitioning over different biome types are not well understood. In this study, we combined machine learning (ML) and causal inference models to investigate and reduce the uncertainties (i.e., parametric, structural, and forcing uncertainties) of CMIP6 simulated evaporative fraction (defined as LE / (LE + H)) across 64 FLUXNET sites that cover five major biomes. We found that CMIP6 model ensembles overestimated evaporative fraction with considerable spread at deciduous broadleaf forest, evergreen needleleaf forest, and savanna sites. Accounting for the biases from all related surface climate and biological driving variables, the CMIP6 model simulated EF could be largely improved (e.g., R between model and observed EF improved from 0.47 to 0.66), with leaf area index, vapor pressure deficit, and precipitation dominated the model improvement. Furthermore, ML-based parameterization generally showed a promising opportunity to further reduce model biases (e.g., R improved from 0.66 to 0.80) in spite of the limited improvement at evergreen broadleaf forest sites where model bias may be dominated by structural imperfection. This study provided an effective framework to understand and reduce model uncertainties in simulating land surface energy flux partitioning and, more importantly, highlighted the need of effective model structure improvement for the next generation earth system land model development.

Highlights • A hierarchical temporal attention based model is proposed to support short-term and long-term human mobility sequence prediction. • The proposed hierarchical temporal attention incorporates individual mobility patterns into the model architecture. • The model is compared with four baseline methods on individual trajectory datasets with varying degree of traveling uncertainty. • Experiments demonstrate the outperformance of the proposed method using three evaluation metrics. • The proposed model uncovers individual frequential and periodical mobility patterns in an interpretable manner., Prediction of individual mobility is crucial in human mobility related applications. Whereas, existing research on individual mobility prediction mainly focuses on next location prediction and short-term dependencies between traveling locations. Long-term location sequence prediction is of great importance for long-time traffic planning and location advertising, and long-term dependencies exist as individual mobility regularity typically occurs daily and weekly. This paper proposes a novel hierarchical temporal attention-based LSTM encoder-decoder model for individual location sequence prediction. The proposed hierarchical attention mechanism captures both long-term and short-term dependencies underlying in individual longitudinal trajectories, and uncovers frequential and periodical mobility patterns in an interpretable manner by incorporating the calendar cycle of individual travel regularities into location prediction. More specifically, the hierarchical attention consists of local temporal attention to identify highly related locations in each day, and global temporal attention to discern important travel regularities over a week. Experiments on individual trajectory datasets with varying degree of traveling uncertainty demonstrate that our method outperforms four baseline methods on three evaluation metrics. In addition, we explore the interpretability of the proposed model in understanding individual daily, and weekly mobility patterns by visualizing the temporal attention weights and frequent traveling patterns associated with locations.

Wildfire is an important ecosystem process, influencing land biogeophysical and biogeochemical dynamics and atmospheric composition. Fire-driven loss of vegetation cover, for example, directly modifies the surface energy budget as a consequence of changing albedo, surface roughness, and partitioning of sensible and latent heat fluxes. Carbon dioxide and methane emitted by fires contribute to a positive atmospheric forcing, whereas emissions of carbonaceous aerosols may contribute to surface cooling. Process-based modeling of wildfires in Earth system land models is challenging due to limited understanding of human, climate, and ecosystem controls on fire counts, fire size, and burned area. Integration of mechanistic wildfire models within Earth system models requires careful parameter calibration, which is computationally expensive and subject to equifinality. To explore alternative approaches, we present a deep neural network (DNN) scheme that surrogates the process-based wildfire model with the Energy Exascale Earth System Model (E3SM) interface. The DNN wildfire model accurately simulates observed burned area with over 90 % higher accuracy with a large reduction in parameterization time compared with the current process-based wildfire model. The surrogate wildfire model successfully captured the observed monthly regional burned area during validation period 2011 to 2015 (coefficient of determination, R2=0.93). Since the DNN wildfire model has the same input and output requirements as the E3SM process-based wildfire model, our results demonstrate the applicability of machine learning for high accuracy and efficient large-scale land model development and predictions.

Wetland CH4 emissions are among the most uncertain components of the global CH4 budget. The complex nature of wetland CH4 processes makes it challenging to identify causal relationships for improving our understanding and predictability of CH4 emissions. In this study, we used the flux measurements of CH4 from eddy covariance towers (30 sites from 4 wetlands types: bog, fen, marsh, and wet tundra) to construct a causality-constrained machine learning (ML) framework to explain the regulative factors and to capture CH4 emissions at sub-seasonal scale. We found that soil temperature is the dominant factor for CH4 emissions in all studied wetland types. Ecosystem respiration (CO2) and gross primary productivity exert controls at bog, fen, and marsh sites with lagged responses of days to weeks. Integrating these asynchronous environmental and biological causal relationships in predictive models significantly improved model performance. More importantly, modeled CH4 emissions differed by up to a factor of 4 under a +1°C warming scenario when causality constraints were considered. These results highlight the significant role of causality in modeling wetland CH4 emissions especially under future warming conditions, while traditional data-driven ML models may reproduce observations for the wrong reasons. Our proposed causality-guided model could benefit predictive modeling, large-scale upscaling, data gap-filling, and surrogate modeling of wetland CH4 emissions within earth system land models.

As a major social problem in Atlanta, one of the cities with high criminal rates in United States, crime has been threatening public security for a long time. Many researches proposed that crime events have characteristics of spatial aggregation and temporal dependence. Therefore, criminal events are predictable, and criminal analysis and prediction is very helpful for law enforcement agencies to more efficiently allocate the limited police forces. In this paper, we use the real crime data set in Atlanta area from 2009 to 2016 to obtain the criminal spatiotemporal distribution features, and make spatiotemporal statistics and visualizations. To predict the daily occurrences of Atlanta crime events, we choose LSTM (long-short term memory) to capture the dependence in time lag and spatial distribution of criminal events. In addition, we discuss the effects of different spatiotemporal scales on the accuracy of crime prediction. When the input time series length is 50 days and the spatial cell size is 0.05 degree, the correlation coefficient R value between the predicted value and the observed records reaches over 0.87. The results provide references for citizens or travelers to avoid hazardous locations, and help law enforcement agencies to allocate resources appropriately.