Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature

Abstract Global warming and urbanization serve as critical research themes in fine-scale climate studies, particularly in developed cities. This study aims to provide a high spatiotemporal resolution dataset of near-surface air temperatures for densely developed urban areas. The dataset comprises da...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengqi Sun, Qingyan Meng, Linlin Zhang, Xinli Hu, Xuewen Lei, Shize Chen, Junyan Hou
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-05032-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849726052903944192
author Mengqi Sun
Qingyan Meng
Linlin Zhang
Xinli Hu
Xuewen Lei
Shize Chen
Junyan Hou
author_facet Mengqi Sun
Qingyan Meng
Linlin Zhang
Xinli Hu
Xuewen Lei
Shize Chen
Junyan Hou
author_sort Mengqi Sun
collection DOAJ
description Abstract Global warming and urbanization serve as critical research themes in fine-scale climate studies, particularly in developed cities. This study aims to provide a high spatiotemporal resolution dataset of near-surface air temperatures for densely developed urban areas. The dataset comprises daily maximum, minimum, and mean temperatures for the summer months (June to August) from 2019 to 2023, at a spatial resolution of 100 m, across the Jiangbei climate zone in China. We applied the Convolutional Long Short-Term Memory (ConvLSTM) deep learning model with multi-source data, including ERA5 temperature data, topography, landcover and vegetation fraction cover. Model evaluation indicates high accuracy, with mean absolute errors (MAE) ranging from 0.564 to 0.784 °C, root mean square errors (RMSE) from 0.733 to 1.027 °C, and coefficients of determination (R2) from 0.892 to 0.943. Our dataset is distinguished by the 100 m spatial resolution and the inclusion of recent summer data from 2023 at a daily scale. This work is valuable for urban or inner-urban climate studies on heatwave mitigation policies and adaptation strategies.
format Article
id doaj-art-ac8bd98453cd406f941085de1d0686aa
institution DOAJ
issn 2052-4463
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Data
spelling doaj-art-ac8bd98453cd406f941085de1d0686aa2025-08-20T03:10:18ZengNature PortfolioScientific Data2052-44632025-05-0112111610.1038/s41597-025-05032-6Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperatureMengqi Sun0Qingyan Meng1Linlin Zhang2Xinli Hu3Xuewen Lei4Shize Chen5Junyan Hou6Aerospace Information Research Institute, Chinese Academy of SciencesAerospace Information Research Institute, Chinese Academy of SciencesAerospace Information Research Institute, Chinese Academy of SciencesAerospace Information Research Institute, Chinese Academy of SciencesAerospace Information Research Institute, Chinese Academy of SciencesAerospace Information Research Institute, Chinese Academy of SciencesBeijing Institute of Remote Sensing InformationAbstract Global warming and urbanization serve as critical research themes in fine-scale climate studies, particularly in developed cities. This study aims to provide a high spatiotemporal resolution dataset of near-surface air temperatures for densely developed urban areas. The dataset comprises daily maximum, minimum, and mean temperatures for the summer months (June to August) from 2019 to 2023, at a spatial resolution of 100 m, across the Jiangbei climate zone in China. We applied the Convolutional Long Short-Term Memory (ConvLSTM) deep learning model with multi-source data, including ERA5 temperature data, topography, landcover and vegetation fraction cover. Model evaluation indicates high accuracy, with mean absolute errors (MAE) ranging from 0.564 to 0.784 °C, root mean square errors (RMSE) from 0.733 to 1.027 °C, and coefficients of determination (R2) from 0.892 to 0.943. Our dataset is distinguished by the 100 m spatial resolution and the inclusion of recent summer data from 2023 at a daily scale. This work is valuable for urban or inner-urban climate studies on heatwave mitigation policies and adaptation strategies.https://doi.org/10.1038/s41597-025-05032-6
spellingShingle Mengqi Sun
Qingyan Meng
Linlin Zhang
Xinli Hu
Xuewen Lei
Shize Chen
Junyan Hou
Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
Scientific Data
title Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
title_full Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
title_fullStr Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
title_full_unstemmed Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
title_short Convolutional Long Short-Term Memory network for generating 100 m daily near-surface air temperature
title_sort convolutional long short term memory network for generating 100 m daily near surface air temperature
url https://doi.org/10.1038/s41597-025-05032-6
work_keys_str_mv AT mengqisun convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT qingyanmeng convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT linlinzhang convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT xinlihu convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT xuewenlei convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT shizechen convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature
AT junyanhou convolutionallongshorttermmemorynetworkforgenerating100mdailynearsurfaceairtemperature