Transformer based models with hierarchical graph representations for enhanced climate forecasting
Abstract Accurate climate predictions are essential for agriculture, urban planning, and disaster management. Traditional forecasting methods often struggle with regional accuracy, computational demands, and scalability. This study proposes a Transformer-based deep learning model for daily temperatu...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-07897-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849238823468269568 |
|---|---|
| author | T. Bhargava Ramu Raviteja Kocherla G. N. V. G. Sirisha V. Lakshmi Chetana P. Vidya Sagar R. Balamurali Nanditha Boddu |
| author_facet | T. Bhargava Ramu Raviteja Kocherla G. N. V. G. Sirisha V. Lakshmi Chetana P. Vidya Sagar R. Balamurali Nanditha Boddu |
| author_sort | T. Bhargava Ramu |
| collection | DOAJ |
| description | Abstract Accurate climate predictions are essential for agriculture, urban planning, and disaster management. Traditional forecasting methods often struggle with regional accuracy, computational demands, and scalability. This study proposes a Transformer-based deep learning model for daily temperature forecasting, utilizing historical climate data from Delhi (2013–2017, consisting of 1,500 daily records). The model integrates three key components: Spatial-Temporal Fusion Module (STFM) to capture spatiotemporal dependencies, Hierarchical Graph Representation and Analysis (HGRA) to model structured climate relationships, and Dynamic Temporal Graph Attention Mechanism (DT-GAM) to enhance temporal feature extraction. To improve computational efficiency and feature selection, we introduce a hybrid optimization approach (HWOA-TTA) that combines the Whale Optimization Algorithm (WOA) and Tiki-Taka Algorithm (TTA). Experimental results demonstrate that the proposed model outperforms baseline models (RF-LSTM-XGBoost, cGAN, CNN + LSTM, and MC-LSTM) by achieving 7.8% higher accuracy, 6.3% improvement in recall, and 8.1% enhancement in F1-score. Additionally, training time is reduced by 22.4% compared to conventional deep learning models, demonstrating improved computational efficiency. These findings highlight the effectiveness of hierarchical graph-based deep learning models for scalable and accurate climate forecasting. Future work will focus on validating the model across diverse climatic regions and enhancing real-time deployment feasibility. |
| format | Article |
| id | doaj-art-18da2d3fada6499bb5ddd23dc9cceff5 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-18da2d3fada6499bb5ddd23dc9cceff52025-08-20T04:01:24ZengNature PortfolioScientific Reports2045-23222025-07-0115112210.1038/s41598-025-07897-4Transformer based models with hierarchical graph representations for enhanced climate forecastingT. Bhargava Ramu0Raviteja Kocherla1G. N. V. G. Sirisha2V. Lakshmi Chetana3P. Vidya Sagar4R. Balamurali5Nanditha Boddu6Department of Electrical and Electronics Engineering, MLR Institute of TechnologyDepartment of Computer Science and Engineering, Malla Reddy UniversityDepartment of CSE, S.R.K.R. Engineering CollegeDepartment of Computer Science and Engineering, School of Computing, Amrita Vishwa VidyapeethamDepartment of Computer Science & Engineering, Koneru Lakshmaiah Education FoundationDepartment of Computer Science and Engineering, Faculty of science and Technology, The ICFAI Foundation for Higher EducationDepartment of Information Technology, Vidya Jyothi Institute of TechnologyAbstract Accurate climate predictions are essential for agriculture, urban planning, and disaster management. Traditional forecasting methods often struggle with regional accuracy, computational demands, and scalability. This study proposes a Transformer-based deep learning model for daily temperature forecasting, utilizing historical climate data from Delhi (2013–2017, consisting of 1,500 daily records). The model integrates three key components: Spatial-Temporal Fusion Module (STFM) to capture spatiotemporal dependencies, Hierarchical Graph Representation and Analysis (HGRA) to model structured climate relationships, and Dynamic Temporal Graph Attention Mechanism (DT-GAM) to enhance temporal feature extraction. To improve computational efficiency and feature selection, we introduce a hybrid optimization approach (HWOA-TTA) that combines the Whale Optimization Algorithm (WOA) and Tiki-Taka Algorithm (TTA). Experimental results demonstrate that the proposed model outperforms baseline models (RF-LSTM-XGBoost, cGAN, CNN + LSTM, and MC-LSTM) by achieving 7.8% higher accuracy, 6.3% improvement in recall, and 8.1% enhancement in F1-score. Additionally, training time is reduced by 22.4% compared to conventional deep learning models, demonstrating improved computational efficiency. These findings highlight the effectiveness of hierarchical graph-based deep learning models for scalable and accurate climate forecasting. Future work will focus on validating the model across diverse climatic regions and enhancing real-time deployment feasibility.https://doi.org/10.1038/s41598-025-07897-4Transformer-based forecastingHierarchical graph modelingClimate predictionFeature optimizationDeep learning |
| spellingShingle | T. Bhargava Ramu Raviteja Kocherla G. N. V. G. Sirisha V. Lakshmi Chetana P. Vidya Sagar R. Balamurali Nanditha Boddu Transformer based models with hierarchical graph representations for enhanced climate forecasting Scientific Reports Transformer-based forecasting Hierarchical graph modeling Climate prediction Feature optimization Deep learning |
| title | Transformer based models with hierarchical graph representations for enhanced climate forecasting |
| title_full | Transformer based models with hierarchical graph representations for enhanced climate forecasting |
| title_fullStr | Transformer based models with hierarchical graph representations for enhanced climate forecasting |
| title_full_unstemmed | Transformer based models with hierarchical graph representations for enhanced climate forecasting |
| title_short | Transformer based models with hierarchical graph representations for enhanced climate forecasting |
| title_sort | transformer based models with hierarchical graph representations for enhanced climate forecasting |
| topic | Transformer-based forecasting Hierarchical graph modeling Climate prediction Feature optimization Deep learning |
| url | https://doi.org/10.1038/s41598-025-07897-4 |
| work_keys_str_mv | AT tbhargavaramu transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT ravitejakocherla transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT gnvgsirisha transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT vlakshmichetana transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT pvidyasagar transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT rbalamurali transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting AT nandithaboddu transformerbasedmodelswithhierarchicalgraphrepresentationsforenhancedclimateforecasting |