Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks

Extreme weather events such as heatwaves, cyclones, floods, wildfires, and droughts are becoming more frequent due to climate change. Climate change causes shifts in biodiversity and impacts agriculture, forest ecosystems, and water resources at a regional scale. However, to study those impacts at t...

Full description

Saved in:
Bibliographic Details
Main Authors: Shailesh Kumar Jha, Vivek Gupta, Priyank J. Sharma, Anurag Mishra, Saksham Joshi
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Climate
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fclim.2025.1572428/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850190091856642048
author Shailesh Kumar Jha
Vivek Gupta
Priyank J. Sharma
Anurag Mishra
Saksham Joshi
author_facet Shailesh Kumar Jha
Vivek Gupta
Priyank J. Sharma
Anurag Mishra
Saksham Joshi
author_sort Shailesh Kumar Jha
collection DOAJ
description Extreme weather events such as heatwaves, cyclones, floods, wildfires, and droughts are becoming more frequent due to climate change. Climate change causes shifts in biodiversity and impacts agriculture, forest ecosystems, and water resources at a regional scale. However, to study those impacts at the regional scale, the spatial resolution provided by the general circulation models (GCMs) and reanalysis products is inadequate. This study evaluates advanced deep learning models for downscaling European Center for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) 2-m temperature data by a factor of 10 (i.e., ranging approximately from 250 to 25 km resolution) for the region spanning 50° to 100° E and 0° to 50° N. We concentrate on gradually improving downscaling models with the help of residual networks. We compare the baseline Super-Resolution Convolutional Neural Network (SRCNN) model with two advanced models: Very Deep Super-Resolution (VDSR) and Enhanced Deep Super-Resolution (EDSR) to assess the impact of residual networks and architectural improvements. The results indicate that VDSR and EDSR significantly outperform SRCNN. Specifically, VDSR increases the Peak Signal-to-Noise Ratio (PSNR) by 4.27 dB and EDSR by 5.23 dB. These models also enhance the Structural Similarity Index Measure (SSIM) by 0.1263 and 0.1163, respectively, indicating better image quality. Furthermore, improvements in the 3°C error threshold are observed, with VDSR and EDSR showing increases of 2.10 and 2.16%, respectively. An explainable artificial intelligence (AI) technique called saliency map analysis provided insights into model performance. Complex terrain areas, such as the Himalayas and the Tibetan Plateau, benefit the most from these advancements. These findings suggest that advanced deep learning models employing residual networks, such as VDSR and EDSR, significantly enhance temperature data accuracy over SRCNN. This approach holds promise for future applications in downscaling other atmospheric variables.
format Article
id doaj-art-c37dfbd9e22a4ee185f8bbcede8d5861
institution OA Journals
issn 2624-9553
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Climate
spelling doaj-art-c37dfbd9e22a4ee185f8bbcede8d58612025-08-20T02:15:24ZengFrontiers Media S.A.Frontiers in Climate2624-95532025-05-01710.3389/fclim.2025.15724281572428Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networksShailesh Kumar Jha0Vivek Gupta1Priyank J. Sharma2Anurag Mishra3Saksham Joshi4School of Civil and Environmental Engineering, Indian Institute of Technology, Mandi, IndiaSchool of Civil and Environmental Engineering, Indian Institute of Technology, Mandi, IndiaDepartment of Civil Engineering, Indian Institute of Technology Indore, Indore, IndiaRegional Remote Sensing Centre - North National Remote Sensing Centre (NRSC), New Delhi, IndiaWater Resources Group, National Remote Sensing Centre (NRSC), Hyderabad, IndiaExtreme weather events such as heatwaves, cyclones, floods, wildfires, and droughts are becoming more frequent due to climate change. Climate change causes shifts in biodiversity and impacts agriculture, forest ecosystems, and water resources at a regional scale. However, to study those impacts at the regional scale, the spatial resolution provided by the general circulation models (GCMs) and reanalysis products is inadequate. This study evaluates advanced deep learning models for downscaling European Center for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) 2-m temperature data by a factor of 10 (i.e., ranging approximately from 250 to 25 km resolution) for the region spanning 50° to 100° E and 0° to 50° N. We concentrate on gradually improving downscaling models with the help of residual networks. We compare the baseline Super-Resolution Convolutional Neural Network (SRCNN) model with two advanced models: Very Deep Super-Resolution (VDSR) and Enhanced Deep Super-Resolution (EDSR) to assess the impact of residual networks and architectural improvements. The results indicate that VDSR and EDSR significantly outperform SRCNN. Specifically, VDSR increases the Peak Signal-to-Noise Ratio (PSNR) by 4.27 dB and EDSR by 5.23 dB. These models also enhance the Structural Similarity Index Measure (SSIM) by 0.1263 and 0.1163, respectively, indicating better image quality. Furthermore, improvements in the 3°C error threshold are observed, with VDSR and EDSR showing increases of 2.10 and 2.16%, respectively. An explainable artificial intelligence (AI) technique called saliency map analysis provided insights into model performance. Complex terrain areas, such as the Himalayas and the Tibetan Plateau, benefit the most from these advancements. These findings suggest that advanced deep learning models employing residual networks, such as VDSR and EDSR, significantly enhance temperature data accuracy over SRCNN. This approach holds promise for future applications in downscaling other atmospheric variables.https://www.frontiersin.org/articles/10.3389/fclim.2025.1572428/fulldownscalingdeep learningtemperatureresidual networksERA5climate change
spellingShingle Shailesh Kumar Jha
Vivek Gupta
Priyank J. Sharma
Anurag Mishra
Saksham Joshi
Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
Frontiers in Climate
downscaling
deep learning
temperature
residual networks
ERA5
climate change
title Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
title_full Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
title_fullStr Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
title_full_unstemmed Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
title_short Deep learning super-resolution for temperature data downscaling: a comprehensive study using residual networks
title_sort deep learning super resolution for temperature data downscaling a comprehensive study using residual networks
topic downscaling
deep learning
temperature
residual networks
ERA5
climate change
url https://www.frontiersin.org/articles/10.3389/fclim.2025.1572428/full
work_keys_str_mv AT shaileshkumarjha deeplearningsuperresolutionfortemperaturedatadownscalingacomprehensivestudyusingresidualnetworks
AT vivekgupta deeplearningsuperresolutionfortemperaturedatadownscalingacomprehensivestudyusingresidualnetworks
AT priyankjsharma deeplearningsuperresolutionfortemperaturedatadownscalingacomprehensivestudyusingresidualnetworks
AT anuragmishra deeplearningsuperresolutionfortemperaturedatadownscalingacomprehensivestudyusingresidualnetworks
AT sakshamjoshi deeplearningsuperresolutionfortemperaturedatadownscalingacomprehensivestudyusingresidualnetworks