English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The accuracy and explainability of data-driven nowcasting models are of greatimportance in many socio-economic sectors reliant on weather-dependent decisionmaking. This paper proposes a novel architecture called Small AttentionResidual UNet (SAR-UNet) for precipitation and cloud cover nowcasting. Here,SmaAt-UNet is used as a core model and is further equipped with residualconnections, parallel to the depthwise separable convolutions. The proposedSAR-UNet model is evaluated on two datasets, i.e., Dutch precipitation mapsranging from 2016 to 2019 and French cloud cover binary images from 2017 to2018. The obtained results show that SAR-UNet outperforms other examined modelsin precipitation nowcasting from 30 to 180 minutes in the future as well ascloud cover nowcasting in the next 90 minutes. Furthermore, we provideadditional insights on the nowcasts made by our proposed model using Grad-CAM,a visual explanation technique, which is employed on different levels of theencoder and decoder paths of the SAR-UNet model and produces heatmapshighlighting the critical regions in the input image as well as intermediaterepresentations to the precipitation. The heatmaps generated by Grad-CAM revealthe interactions between the residual connections and the depthwise separableconvolutions inside of the multiple depthwise separable blocks placedthroughout the network architecture.

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