<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>This study delves into the synergy between remote sensing and satellite gravimetry, focusing on the utilization of Advanced Microwave Scanning Radiometer (AMSR-E) data for modeling delta Total Water Storage (ΔTWS) values derived from the GRACE mission. Various machine learning algorithms were employed to investigate the concordance between Gravity Recovery and Climate Experiment (GRACE) and AMSR-E observations. Despite the limited correlation in circumpolar permafrost areas, ΔTWS was successfully modeled with an accuracy of a Root Mean Square Error (RMSE) of 3.5 cm. The Amazon region exhibited a notable model error, attributed to significant ΔTWS amplitude; the overall model quality was affirmed by Normalized Root Mean Square Error (NRMSE) and Nash-Sutcliffe Efficiency (NSE) metrics. Importantly, the effectiveness of AMSR-E Soil Moisture (SM) data, encompassing C (frequency of 4–8 GHz) and X (frequency of 8–12 GHz) ranges (~0.04 m and ~0.03 m wavelength, respectively) in modeling ΔTWS, even in heavily forested equatorial regions, was demonstrated.</p>
</abstract>

This study delves into the synergy between remote sensing and satellite gravimetry, focusing on the utilization of Advanced Microwave Scanning Radiometer (AMSR-E) data for modeling delta Total Water Storage (ΔTWS) values derived from the GRACE mission. Various machine learning algorithms were employed to investigate the concordance between Gravity Recovery and Climate Experiment (GRACE) and AMSR-E observations. Despite the limited correlation in circumpolar permafrost areas, ΔTWS was successfully modeled with an accuracy of a Root Mean Square Error (RMSE) of 3.5 cm. The Amazon region exhibited a notable model error, attributed to significant ΔTWS amplitude; the overall model quality was affirmed by Normalized Root Mean Square Error (NRMSE) and Nash-Sutcliffe Efficiency (NSE) metrics. Importantly, the effectiveness of AMSR-E Soil Moisture (SM) data, encompassing C (frequency of 4–8 GHz) and X (frequency of 8–12 GHz) ranges (~0.04 m and ~0.03 m wavelength, respectively) in modeling ΔTWS, even in heavily forested equatorial regions, was demonstrated.

<bold>Using machine learning techniques</bold> to reconstruct the signal observed by the GRACE mission based on AMSR-E microwave data

Department of Geodesy and Geodetic Astronomy, Faculty of Geodesy and Cartography

Department of Photogrammetry, Remote Sensing and Spatial Information Systems, Faculty of Geodesy and Cartography

Miscellanea Geographica

This work is licensed under the Creative Commons Attribution 4.0 International License.

 to reconstruct the signal observed by the GRACE mission based on AMSR-E microwave data

This study delves into the synergy between remote sensing and satellite gravimetry, focusing on the utilization of Advanced Microwave Scanning Radiometer...

Model	RMSE [m]	R²	Δ RMSE [%]	Δ R² [%]	1-R²	Δ 1-R² [%]
Random Forest Regressor	0.035	0.761	51.3	380700.0	0.239	76.1
Extra Trees Regressor	0.035	0.757	50.9	378700.0	0.243	75.7
Extreme Gradient Boosting	0.037	0.739	48.9	369350.0	0.262	73.9
K Neighbors Regressor	0.038	0.725	47.7	362450.0	0.275	72.5
Light Gradient Boosting Machine	0.038	0.715	46.7	357750.0	0.285	71.5
Decision Tree Regressor	0.048	0.546	32.8	273000.0	0.454	54.6
Gradient Boosting Regressor	0.052	0.469	27.3	234600.0	0.531	46.9
Linear Regression	0.069	0.074	3.9	36950.0	0.926	7.4
Least Angle Regression	0.069	0.074	3.9	36950.0	0.926	7.4
Bayesian Ridge	0.069	0.074	3.9	36900.0	0.926	7.4
Ridge Regression	0.069	0.068	3.7	34150.0	0.932	6.8
Huber Regressor	0.070	0.062	3.2	31000.0	0.938	6.2
Orthogonal Matching Pursuit	0.072	0.000	0.2	50.0	1.000	0.0
Lasso Regression	0.072	−0.001	0.2	−150.0	1.001	0.0
Elastic Net	0.072	−0.001	0.2	−150.0	1.001	0.0
Lasso Least Angle Regression	0.072	−0.001	0.2	−150.0	1.001	0.0
Dummy Regressor	0.072	−0.001	0.2	−150.0	1.001	0.0
AdaBoost Regressor	0.073	−0.021	−0.9	−10450.0	1.021	−2.1
Passive Aggressive Regressor	0.086	−0.485	−20.0	−242550.0	1.485	−48.5
sin+cos annual function (baseline)	0.072	0.000	-	-	1.000
sin+cos semiannual function	0.095	0.000	−32.7	0.0	1.000	0.0

Using machine learning techniques to reconstruct the signal observed by the GRACE mission based on AMSR-E microwave data

Data publikacji: 30 kwi 2024

Zakres stron: 80 - 86

Otrzymano: 20 sty 2024

Przyjęty: 21 kwi 2024

DOI: https://doi.org/10.2478/mgrsd-2023-0033

Słowa kluczowe
GRACE, AMSR-E, total water storage anomalies, soil moisture, remote sensors

© 2024 Viktor Szabó et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1.

Figure 2.

Figure 3.

The achieved results on the test data sample

Using machine learning techniques to reconstruct the signal observed by the GRACE mission based on AMSR-E microwave data

Data publikacji: 30 kwi 2024

Zakres stron: 80 - 86

Otrzymano: 20 sty 2024

Przyjęty: 21 kwi 2024

DOI: https://doi.org/10.2478/mgrsd-2023-0033

Słowa kluczoweGRACE, AMSR-E, total water storage anomalies, soil moisture, remote sensors

© 2024 Viktor Szabó et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1.

Figure 2.

Figure 3.

The achieved results on the test data sample

Słowa kluczowe
GRACE, AMSR-E, total water storage anomalies, soil moisture, remote sensors