Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada
Estimating terrestrial water storage (TWS) with high spatial resolution is crucial for hydrological and water resource management. Comparing to traditional in-situ data measurement, observation from space borne sensor such as Gravity Recovery and Climate Experiment (GRACE) satellites is quite effect...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2021-07-01
|
| Series: | Canadian Journal of Remote Sensing |
| Online Access: | http://dx.doi.org/10.1080/07038992.2021.1954498 |
| _version_ | 1797661099945885696 |
|---|---|
| author | Hongjie He Ke Yang Shusen Wang Hasti Andon Petrosians Ming Liu Junhua Li José Marcato Junior Wesley Nunes Gonçalves Lanying Wang Jonathan Li |
| author_facet | Hongjie He Ke Yang Shusen Wang Hasti Andon Petrosians Ming Liu Junhua Li José Marcato Junior Wesley Nunes Gonçalves Lanying Wang Jonathan Li |
| author_sort | Hongjie He |
| collection | DOAJ |
| description | Estimating terrestrial water storage (TWS) with high spatial resolution is crucial for hydrological and water resource management. Comparing to traditional in-situ data measurement, observation from space borne sensor such as Gravity Recovery and Climate Experiment (GRACE) satellites is quite effective to obtain a large-scale TWS data. However, the coarse resolution of the GRACE data restricts its application at a local scale. This paper presents three novel convolutional neural network (CNN) based approaches including the Super-Resolution CNN (SRCNN), the Very Deep Super-Resolution (VDSR), and the Residual Channel Attention Networks (RCAN) to spatial downscaling of the monthly GRACE TWS products using the outputs of the Ecological Assimilation of Land and Climate Observations (EALCO) model over Canada. We also compare the performance of CNN-based methods with the empirical linear regression-based downscaling method. All comparison results were evaluated by root mean square error (RMSE) between the reconstructed GRACE TWS and the original one. RMSEs over the matched pixels are 22.3, 14.4, 18.4 and 71.6 mm of SRCNN, VDSR, RCAN and linear regression-based method respectively. Obviously, VDSR shows the best accuracy among all methods. The result shows all CNN-based super resolution methods preform much better than traditional method in spatial downscaling. |
| first_indexed | 2024-03-11T18:40:25Z |
| format | Article |
| id | doaj.art-1bf879e95de54c2c92eefbe7cb206b91 |
| institution | Directory Open Access Journal |
| issn | 1712-7971 |
| language | English |
| last_indexed | 2024-03-11T18:40:25Z |
| publishDate | 2021-07-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Canadian Journal of Remote Sensing |
| spelling | doaj.art-1bf879e95de54c2c92eefbe7cb206b912023-10-12T13:36:24ZengTaylor & Francis GroupCanadian Journal of Remote Sensing1712-79712021-07-0147465767510.1080/07038992.2021.19544981954498Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over CanadaHongjie He0Ke Yang1Shusen Wang2Hasti Andon Petrosians3Ming Liu4Junhua Li5José Marcato Junior6Wesley Nunes Gonçalves7Lanying Wang8Jonathan Li9Department of Geography and Environmental Management, University of WaterlooDepartment of Systems Design Engineering, University of WaterlooCanada Centre for Remote Sensing, Natural Resources CanadaDepartment of Geography and Environmental Management, University of WaterlooDepartment of Geography and Environmental Management, University of WaterlooCanada Centre for Remote Sensing, Natural Resources CanadaFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do SulFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do SulDepartment of Geography and Environmental Management, University of WaterlooDepartment of Geography and Environmental Management, University of WaterlooEstimating terrestrial water storage (TWS) with high spatial resolution is crucial for hydrological and water resource management. Comparing to traditional in-situ data measurement, observation from space borne sensor such as Gravity Recovery and Climate Experiment (GRACE) satellites is quite effective to obtain a large-scale TWS data. However, the coarse resolution of the GRACE data restricts its application at a local scale. This paper presents three novel convolutional neural network (CNN) based approaches including the Super-Resolution CNN (SRCNN), the Very Deep Super-Resolution (VDSR), and the Residual Channel Attention Networks (RCAN) to spatial downscaling of the monthly GRACE TWS products using the outputs of the Ecological Assimilation of Land and Climate Observations (EALCO) model over Canada. We also compare the performance of CNN-based methods with the empirical linear regression-based downscaling method. All comparison results were evaluated by root mean square error (RMSE) between the reconstructed GRACE TWS and the original one. RMSEs over the matched pixels are 22.3, 14.4, 18.4 and 71.6 mm of SRCNN, VDSR, RCAN and linear regression-based method respectively. Obviously, VDSR shows the best accuracy among all methods. The result shows all CNN-based super resolution methods preform much better than traditional method in spatial downscaling.http://dx.doi.org/10.1080/07038992.2021.1954498 |
| spellingShingle | Hongjie He Ke Yang Shusen Wang Hasti Andon Petrosians Ming Liu Junhua Li José Marcato Junior Wesley Nunes Gonçalves Lanying Wang Jonathan Li Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada Canadian Journal of Remote Sensing |
| title | Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada |
| title_full | Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada |
| title_fullStr | Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada |
| title_full_unstemmed | Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada |
| title_short | Deep Learning Approaches to Spatial Downscaling of GRACE Terrestrial Water Storage Products Using EALCO Model Over Canada |
| title_sort | deep learning approaches to spatial downscaling of grace terrestrial water storage products using ealco model over canada |
| url | http://dx.doi.org/10.1080/07038992.2021.1954498 |
| work_keys_str_mv | AT hongjiehe deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT keyang deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT shusenwang deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT hastiandonpetrosians deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT mingliu deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT junhuali deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT josemarcatojunior deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT wesleynunesgoncalves deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT lanyingwang deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada AT jonathanli deeplearningapproachestospatialdownscalingofgraceterrestrialwaterstorageproductsusingealcomodelovercanada |