Improving Atmospheric Angular Momentum Forecasts by Machine Learning

Dill, Robert; Saynisch, Jan; Irrgang, Christopher; Thomas, Maik

doi:10.1029/2021ea002070

Cited by 7 publications

(8 citation statements)

References 26 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An important point regarding the results in Table 2 is that the OAM functions are the most important ones for the prediction of polar motion and this is in agreement with the results of Dill et al. (2021). Dahlen (1976) discovered that oceans can increase the main period of polar motion by around 28 days.…”

Section: Discussionsupporting

confidence: 89%

“…AAM per se does not result in a good performance. This attests to the fact that AAM, when used alone, is not a helpful feature to be used for the prediction of polar motion (Dill et al., 2021). Furthermore, combinations in which AAM is used alongside the OAM show better performance compared to the case of AAM alone, possibly due to the correlation between AAM and OAM.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Neural ODE Differential Learning and Its Application in Polar Motion Prediction

2022

View full text Add to dashboard Cite

This paper introduces a new learning algorithm for accurate, physically driven time series prediction. The fundamental assumption behind the method is that the phenomena follow Ordinary Differential Equations. We investigate the general case where the time series follows an ODE of degree m∈double-struckN $m\in \mathbb{N}$. The resulting method is a learning algorithm based on the finite differences between the values of time series. We present the application of the method in the field of geodesy for polar motion prediction, the main objective of the present paper. We show that in this application, the linear form of the method is sufficient and offers competitive predictive performance. We present a baseline solution, in which we use historical polar motion time series from 1976 to predict up to the year 2020. The prediction horizon in this case is short‐term (up to 10 days into the future). In addition, we compare the prediction accuracy in the short‐term horizon with the results of the best performing model in the first Earth Orientation Prediction Comparison Campaign. On average, a 53% improvement in prediction performance is achieved. In further analyses, we compare the prediction accuracy for both short‐term and long‐term against the results of state‐of‐the‐art methods, namely Multichannel Singular Spectrum Analysis, and a combination of Singular Spectrum Analysis and Copula sampling. We show that the proposed method in this paper can outperform the mentioned two methods in both short and long‐term horizons, with an average improvement of the prediction performance of 54% and 52%, respectively.

show abstract

Section: Discussionsupporting

confidence: 89%

Section: Discussionmentioning

confidence: 99%

Neural ODE Differential Learning and Its Application in Polar Motion Prediction

2022

View full text Add to dashboard Cite

show abstract

“…There have been successful applications of machine learning for the analysis and prediction of EOPs (Dill et al., 2021; Gou et al., 2023; Kiani Shahvandi & Soja, 2022a, 2022b; Kiani Shahvandi, Schartner, & Soja, 2022). Here, however, we need to consider the specific aspects of the problem and develop a new machine learning algorithm.…”

Section: Introductionmentioning

confidence: 99%

Geophysically Informed Machine Learning for Improving Rapid Estimation and Short‐Term Prediction of Earth Orientation Parameters

Kiani Shahvandi,

Dill,

Dobslaw

et al. 2023

JGR Solid Earth

Self Cite

View full text Add to dashboard Cite

Rapid provision of Earth Orientation Parameters (EOPs, here polar motion and dUT1) is indispensable in many geodetic applications and also for spacecraft navigation. There are, however, discrepancies between the rapid EOPs and the final EOPs that have a higher latency, but the highest accuracy. To reduce these discrepancies, we focus on a data‐driven approach, present a novel method named ResLearner, and use it in the context of deep ensemble learning. Furthermore, we introduce a geophysically‐constrained approach for ResLearner. We show that the most important geophysical information to improve the rapid EOPs is the effective angular momentum functions of atmosphere, ocean, land hydrology, and sea level. In addition, semi‐diurnal, diurnal, and long‐period tides coupled with prograde and retrograde tidal excitations are important features. The influence of some climatic indices on the prediction accuracy of dUT1 is discussed and El Niño Southern Oscillation is found to be influential. We developed an operational framework, providing the improved EOPs on a daily basis with a prediction window of 63 days to fully cover the latency of final EOPs. We show that under the operational conditions and using the rapid EOPs of the International Earth Rotation and Reference Systems Service (IERS) we achieve improvements as high as 60%, thus significantly reducing the differences between rapid and final EOPs. Furthermore, we discuss how the new final series IERS 20 C04 is preferred over 14 C04. Finally, we compare against EOP hindcast experiments of the European Space Agency, on which ResLearner presents comparable improvements.

show abstract

“…All prediction techniques that show as low mean absolute errors as the official prediction of the International Earth Rotation and Reference Systems Service (IERS), called bulletin A, or even better use at least AAM forecasts or all available EAM forecasts (Kur et al, 2022). Best prediction methods also implicitly consider the EAM forecast errors by a combination of AAM forecasts with AAM analysis data (Dill et al, 2019), elimination of AAM wind term forecast errors by a cascade forward neural network (Dill et al, 2021), or generating improved EAM forecasts using machine learning (Kiani et al, 2022b;Gou et al, 2023). Although such new methods can reduce the influence of existing EAM forecast errors, a more detailed picture of the EAM forecast uncertainties is hidden in such approaches.…”

Section: Introductionmentioning

confidence: 99%

ESMGFZ EAM Products for EOP Prediction: Toward the Quantification of Time Variable EAM Forecast Errors

Dill,

Dobslaw,

Thomas

2023

Artificial Satellites

View full text Add to dashboard Cite

Since more than 10 years, the Earth system modeling group at GFZ (ESMGFZ) provides effective angular momentum (EAM) functions for Earth orientation parameter assessment on a routinely daily basis. In addition to EAM of the individual Earth’s subsystems atmosphere, ocean, and hydrology, the global mass balance is calculated as barystatic sea level variation by solving explicitly the sea-level equation. ESMGFZ provides also 6-day forecasts for all of these EAM products. EAM forecasts are naturally degraded by forecast errors that typically grow with increasing forecast length, but they also show recurring patterns with occasionally higher errors at very short forecast horizons. To characterize such errors which are not randomly distributed in time, we divided the errors into a systematic and a stochastic contribution. In an earlier study, we were able to detect and remove the large systematic fraction occurring in the atmospheric angular momentum (AAM) wind term forecast errors with a cascading forward neural network model, thereby reducing the total forecast error by about 50%. In contrast, we were not able to remove the random error component assed in this study. Nevertheless, we show that machine learning methods are able to predict quasi-daily variations in time variable EAM forecasts error levels. We plan to provide these forecast error estimates along with the deterministic EAM forecast products for subsequent use in, for example, EOP Kalman filter prediction schemes.

show abstract

Improving Atmospheric Angular Momentum Forecasts by Machine Learning

Cited by 7 publications

References 26 publications

Neural ODE Differential Learning and Its Application in Polar Motion Prediction

Neural ODE Differential Learning and Its Application in Polar Motion Prediction

Geophysically Informed Machine Learning for Improving Rapid Estimation and Short‐Term Prediction of Earth Orientation Parameters

ESMGFZ EAM Products for EOP Prediction: Toward the Quantification of Time Variable EAM Forecast Errors

Contact Info

Product

Resources

About