# SAR-UNet: Small Attention Residual UNet for Explainable Nowcasting Tasks

arXiv: https://arxiv.org/abs/2303.06663 · HTML: https://mlsanity.com/paper/2303.06663/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 34 references that are in the map.
- First posted on arXiv 12 Mar 2023.

## Builds on (most cited first, 15 of 15)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data](https://mlsanity.com/paper/1904.00592/) (2019)
- [Explainable Deep Learning: A Field Guide for the Uninitiated](https://mlsanity.com/paper/2004.14545/) (2020)
- [SmaAt-UNet: Precipitation Nowcasting using a Small Attention-UNet Architecture](https://mlsanity.com/paper/2007.04417/) (2020)
- [Broad-UNet: Multi-scale feature learning for nowcasting tasks](https://mlsanity.com/paper/2102.06442/) (2021)
- [AA-TransUNet: Attention Augmented TransUNet For Nowcasting Tasks](https://mlsanity.com/paper/2202.04996/) (2022)
- [Deep Graph Convolutional Networks for Wind Speed Prediction](https://mlsanity.com/paper/2101.10041/) (2021)
- [Deep coastal sea elements forecasting using U-Net based models](https://mlsanity.com/paper/2011.03303/) (2020)
- [Multistream Graph Attention Networks for Wind Speed Forecasting](https://mlsanity.com/paper/2108.07063/) (2021)
- [Deep multi-stations weather forecasting: explainable recurrent convolutional neural networks](https://mlsanity.com/paper/2009.11239/) (2020)

## Built on by (most cited first, 6 of 6)

- [GA-SmaAt-GNet: Generative Adversarial Small Attention GNet for Extreme Precipitation Nowcasting](https://mlsanity.com/paper/2401.09881/) (2024)
- [Interpretable Machine Learning for Weather and Climate Prediction: A Survey](https://mlsanity.com/paper/2403.18864/) (2024)
- [SSA-UNet: Advanced Precipitation Nowcasting via Channel Shuffling](https://mlsanity.com/paper/2504.18309/) (2025)
- [Self-supervised Spatial-Temporal Learner for Precipitation Nowcasting](https://mlsanity.com/paper/2412.15917/) (2024)
- [Integrating Weather Station Data and Radar for Precipitation Nowcasting: SmaAt-fUsion and SmaAt-Krige-GNet](https://mlsanity.com/paper/2502.16116/) (2025)
- [SmaAT-QMix-UNet: A Parameter-Efficient Vector-Quantized UNet for Precipitation Nowcasting](https://mlsanity.com/paper/2603.21879/) (2026)

## Abstract (arXiv)

The accuracy and explainability of data-driven nowcasting models are of great importance in many socio-economic sectors reliant on weather-dependent decision making. This paper proposes a novel architecture called Small Attention Residual UNet (SAR-UNet) for precipitation and cloud cover nowcasting. Here, SmaAt-UNet is used as a core model and is further equipped with residual connections, parallel to the depthwise separable convolutions. The proposed SAR-UNet model is evaluated on two datasets, i.e., Dutch precipitation maps ranging from 2016 to 2019 and French cloud cover binary images from 2017 to 2018. The obtained results show that SAR-UNet outperforms other examined models in precipitation nowcasting from 30 to 180 minutes in the future as well as cloud cover nowcasting in the next 90 minutes. Furthermore, we provide additional insights on the nowcasts made by our proposed model using Grad-CAM, a visual explanation technique, which is employed on different levels of the encoder and decoder paths of the SAR-UNet model and produces heatmaps highlighting the critical regions in the input image as well as intermediate representations to the precipitation. The heatmaps generated by Grad-CAM reveal the interactions between the residual connections and the depthwise separable convolutions inside of the multiple depthwise separable blocks placed throughout the network architecture.