IMPA-Net: Meteorology-Aware Multi-Scale Attention and Dynamic Loss for Extreme Convective Radar Nowcasting

• URL: http://arxiv.org/abs/2604.24224v1
• Date: Mon, 27 Apr 2026 09:30:07 GMT
• Title: IMPA-Net: Meteorology-Aware Multi-Scale Attention and Dynamic Loss for Extreme Convective Radar Nowcasting
• Abstract summary: Short-range prediction of convective precipitation from weather radar observations is essential for severe weather warnings.<n>We propose IMPA-Net, a deterministic 0-2 hour nowcasting framework that addresses limitations through meteorologically-informed designs at the input, architecture, and loss function levels.
• Score: 2.405586180193259
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Short-range prediction of convective precipitation from weather radar observations is essential for severe weather warnings. However, deep learning models trained with pixel-wise error metrics tend to produce overly smooth forecasts that suppress intense echoes critical for hazard detection. This issue is exacerbated by insufficient multi-scale feature interaction and suboptimal fusion of heterogeneous geophysical inputs. We propose IMPA-Net (Integrated Multi-scale Predictive Attention Network), a deterministic 0-2 hour nowcasting framework that addresses these limitations through meteorologically-informed designs at the input, architecture, and loss function levels. A parameter-free Spatial Mixer reorganizes heterogeneous input channels at the mesoscale-$γ$ neighborhood (~2 km) via deterministic channel permutation, providing a structured cross-field prior. An integrated multi-scale predictive attention module serves as the spatiotemporal translator, capturing dynamics from mesoscale-$β$ to mesoscale-$γ$ scales. A Meteorologically-Aware Dynamic Loss employs three-level asymmetric weighting -- adapting across training epochs, storm intensity, and forecast lead time -- to counteract regression-to-the-mean. Evaluated against seven baselines on a multi-source radar dataset over eastern China, IMPA-Net raises the Heidke Skill Score at $\geq$45 dBZ from 0.049 (SimVP baseline) to 0.143 under matched settings. Relative to pySTEPS, it provides a better trade-off between severe-event detection and false-alarm control. Spectral analysis confirms preserved energy across mesoscale bands where competing methods show progressive smoothing. These improvements are shown within a single domain and convective regime; generalizability to other orographic and climatic regions remains to be tested.

Related papers

• Physics-Based Deep Spatiotemporal Hyperlocal Radar Nowcasting with a Multi-Variable U-Net for High-Resolution Precipitation Forecasting [0.0]
  We develop a radar-only nowcasting framework that combines multi-elevation reflectivity, Doppler radial velocity, and radial-velocity-gradient proxy features within an encoder-decoder U-Net.<n>The model is trained using Mumbai Doppler radar observations from May to August 2023 and evaluated on temporally independent events.
  arXiv  Detail & Related papers  (2026-07-17T16:09:48Z)
• Physics-Guided Dual Decoding and Spectral Supervision for Global 3D Hydrometeor Prediction [2.1603484118769476]
  Three-dimensional hydrometeor forecasting is challenging due to the zero-inflated, long-tailed distributions of these variables.<n>We propose PredHydro-Net, a physics-guided dual-decoding framework that mitigates this smoothing.<n>By integrating wavelet-based frequency decoupling, spectral amplitude matching, and adversarial training, the model achieves a favorable trade-off between quantitative accuracy and spatial fidelity.
  arXiv  Detail & Related papers  (2026-06-07T10:38:37Z)
• PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting [62.22445401483844]
  Existing methods are severely constrained by the bottleneck of compounding errors.<n>We propose a universal framework called -Cast-Corrector (Plug-and-Play Corrector)<n>Our method significantly enhances the long-term stability of coupled forecasting systems.
  arXiv  Detail & Related papers  (2026-05-09T13:12:33Z)
• Enhancing AI-Based Tropical Cyclone Track and Intensity Forecasting via Systematic Bias Correction [10.587035638034767]
  Tropical cyclones pose severe threats to life, infrastructure, and economies in tropical and subtropical regions.<n>Recent advances in AI-based weather forecasting have shown promise in improving track forecasts.<n>We propose BaguanCyclone, a novel, unified framework that integrates two key innovations.
  arXiv  Detail & Related papers  (2026-03-20T03:25:03Z)
• AGCD: Agent-Guided Cross-Modal Decoding for Weather Forecasting [29.582978733703687]
  Agent-Guided Cross-modal Decoding (AGCD) is a plug-and-play decoding-time prior-injection paradigm.<n>We design a multi-agent meteorological narration pipeline to generate state-conditioned physics-priors.<n>We also introduce cross-modal region interaction decoding that performs region-aware multi-scale tokenization and efficient physics-priors injection.
  arXiv  Detail & Related papers  (2026-03-16T13:27:20Z)
• MMWSTM-ADRAN+: A Novel Hybrid Deep Learning Architecture for Enhanced Climate Time Series Forecasting and Extreme Event Prediction [0.0]
  We present Multi-Modal Weather State Transition Model with Anomaly-Driven Recurrent Attention Network Plus (MMWSTM-ADRAN+)<n> MMWSTM combines bidirectional Long Short-Term Memory (BiLSTM) units with a learnable Markov state transition matrix to capture synoptic-scale weather regime changes.<n>The second stream, ADRAN, integrates bidirectional Gated Recurrent Units (BiGRUs), multi-head self-attention, and a novel anomaly amplification layer to enhance sensitivity to low-probability signals.
  arXiv  Detail & Related papers  (2025-11-17T14:31:54Z)
• CSU-PCAST: A Dual-Branch Transformer Framework for medium-range ensemble Precipitation Forecasting [6.540270371082014]
  This study develops a deep learning-based ensemble framework for multi-step precipitation prediction.<n>The architecture employs a patch-based Swin Transformer backbone with periodic convolutions to handle longitudinal continuity.<n>Training minimizes a hybrid loss combining the Continuous Ranked Probability Score (CRPS) and weighted log1p mean squared error (log1pMSE)
  arXiv  Detail & Related papers  (2025-10-23T17:43:38Z)
• SynCast: Synergizing Contradictions in Precipitation Nowcasting via Diffusion Sequential Preference Optimization [62.958457694151384]
  We introduce preference optimization into precipitation nowcasting for the first time, motivated by the success of reinforcement learning from human feedback in large language models.<n>In the first stage, the framework focuses on reducing FAR, training the model to effectively suppress false alarms.
  arXiv  Detail & Related papers  (2025-10-22T16:11:22Z)
• S$^2$Transformer: Scalable Structured Transformers for Global Station Weather Forecasting [67.93713728260646]
  Existing time series forecasting methods often ignore or unidirectionally model spatial correlation when conducting large-scale global station forecasting.<n>This contradicts the nature underlying observations of the global weather system limiting forecast performance.<n>We propose a novel Structured Spatial Attention in this paper.<n>It partitions the spatial graph into a set of subgraphs and instantiates Intra-subgraph Attention to learn local spatial correlation within each subgraph.<n>It aggregates nodes into subgraph representations for message passing among the subgraphs via Inter-subgraph Attention -- considering both spatial proximity and global correlation.
  arXiv  Detail & Related papers  (2025-09-10T05:33:28Z)
• EMForecaster: A Deep Learning Framework for Time Series Forecasting in Wireless Networks with Distribution-Free Uncertainty Quantification [11.88654374102813]
  We develop a deep learning (DL) time series forecasting framework referred to as textitEMForecaster.<n> EMForecaster achieves superior performance across diverse datasets, spanning both short-term and long-term prediction horizons.<n>In point forecasting tasks, EMForecaster substantially outperforms current state-of-the-art DL approaches, showing improvements of 53.97% over the Transformer architecture and 38.44% over the average of all baseline models.
  arXiv  Detail & Related papers  (2025-03-31T18:10:08Z)
• Generating Fine-Grained Causality in Climate Time Series Data for Forecasting and Anomaly Detection [67.40407388422514]
  We design a conceptual fine-grained causal model named TBN Granger Causality. Second, we propose an end-to-end deep generative model called TacSas, which discovers TBN Granger Causality in a generative manner. We test TacSas on climate benchmark ERA5 for climate forecasting and the extreme weather benchmark of NOAA for extreme weather alerts.
  arXiv  Detail & Related papers  (2024-08-08T06:47:21Z)
• Observation-Guided Meteorological Field Downscaling at Station Scale: A Benchmark and a New Method [66.80344502790231]
  We extend meteorological downscaling to arbitrary scattered station scales and establish a new benchmark and dataset. Inspired by data assimilation techniques, we integrate observational data into the downscaling process, providing multi-scale observational priors. Our proposed method outperforms other specially designed baseline models on multiple surface variables.
  arXiv  Detail & Related papers  (2024-01-22T14:02:56Z)
• Learning Robust Precipitation Forecaster by Temporal Frame Interpolation [65.5045412005064]
  We develop a robust precipitation forecasting model that demonstrates resilience against spatial-temporal discrepancies. Our approach has led to significant improvements in forecasting precision, culminating in our model securing textit1st place in the transfer learning leaderboard of the textitWeather4cast'23 competition.
  arXiv  Detail & Related papers  (2023-11-30T08:22:08Z)
• Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling [58.456404022536425]
  State of the art for physical hazard prediction from weather and climate requires expensive km-scale numerical simulations driven by coarser resolution global inputs. Here, a generative diffusion architecture is explored for downscaling such global inputs to km-scale, as a cost-effective machine learning alternative. The model is trained to predict 2km data from a regional weather model over Taiwan, conditioned on a 25km global reanalysis.
  arXiv  Detail & Related papers  (2023-09-24T19:57:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.