Register now After registration you will be able to apply for this opportunity online.
Enhancing the spatial resolution of monthly gravity solutions derived from satellite laser ranging
The variations of the Earth's gravity field on a monthly scale mainly indicate the global water dynamics and provide invaluable information for understanding the Earth system[1]. Monthly gravity variations can be measured by GRACE(-FO) satellite missions with higher accuracy, but the data availability of such satellite missions is limited. On the other hand, the monthly gravity field variations can also be detected by analyzing orbit perturbations of satellites in low-Earth orbit (LEO)[2], or using the satellite laser ranging (SLR)[3,4] techniques. SLR provides extended data records but typically with worse spatial resolution and accuracy.
Keywords: Deep learning, GRACE-FO, LEO, SLR
Group work: yes
English
Group work: yes
English
In this study, the student should design a deep learning model to generate GRACE-like monthly solutions from the low-resolution SLR dataset. The student should first perform a literature review to get familiar with the monthly gravity field observations and products. Then, the relevant input features should be defined, and possible deep learning algorithms, such as generative models, should be designed. The model will be trained with available GRACE(-FO) data as reference and then applied to the epochs without GRACE(-FO) products. The benefit of final predicted monthly gravity field variations will be evaluated from hydrological perspectives.
In this study, the student should design a deep learning model to generate GRACE-like monthly solutions from the low-resolution SLR dataset. The student should first perform a literature review to get familiar with the monthly gravity field observations and products. Then, the relevant input features should be defined, and possible deep learning algorithms, such as generative models, should be designed. The model will be trained with available GRACE(-FO) data as reference and then applied to the epochs without GRACE(-FO) products. The benefit of final predicted monthly gravity field variations will be evaluated from hydrological perspectives.
Junyang Gou (jungou@ethz.ch)
Prof. Dr. Benedikt Soja (soja@ethz.ch)
Junyang Gou (jungou@ethz.ch) Prof. Dr. Benedikt Soja (soja@ethz.ch)
Enhancing the spatial resolution of monthly gravity solutions derived from satellite laser ranging.pdf
