Table 5 True values, search space, and inversion results for dual anomaly case using EKI with noise-free and 10% Gaussian noise-contaminated data.

From: Regularized ensemble Kalman inversion for robust and efficient gravity data modeling to identify mineral and ore deposits

Model parameters

True

Search space

Noise free

Noise contaminated

\({A_1}\left( {{\text{~mGal}}.{{\text{m}}^2}{\text{~}}} \right)\)

600

400–700

639.22 ± 18.70

640.45 ± 36.21

\({z_1}\left( m \right)\)

20

10–40

19.73 ± 0.38

20.98 ± 0.44

\({x_{01}}\left( m \right)\)

-150

-200–200

-149.99 ± 0.01

-149.77 ± 0.05

\({q_1}\)

1.5

0.5–1.5

1.47 ± 0.03

1.49 ± 0.13

\({\mu _1}\)

1

0–1

0.91 ± 0.08

0.99 ± 0.29

\({A_2}\left( {{\text{~mGal}}.{{\text{m}}^2}{\text{~}}} \right)\)

150

50–300

212.66 ± 19.89

247.07 ± 20.56

\({z_2}\left( {\text{m}} \right)\)

40

20–80

40.04 ± 0.04

42.32 ± 0.27

\({x_{02}}\left( m \right)\)

100

-200–200

99.99 ± 0.00

100.36 ± 0.03

\({q_2}\)

1

0.5–1.5

1.00 ± 0.00

1.04 ± 0.00

\({\mu _2}\)

1

0–1

0.90 ± 0.02

0.96 ± 0.03

Minimum objective function (mGal)

7.68e-05

0.14

The median objective function (mGal)

2.48e-04

0.14

The interquartile range of the objective function (mGal)

1.10e-04

1.29e-05