A regional geomagnetic field model for China based on Swarm satellite data and 3D Legendre polynomials

doi:10.11947/j.AGCS.2025.20240379

Abstract

Abstract:

Regional geomagnetic field models can provide detailed information about the geomagnetic field, with significant applications in precise navigation and target detection. To establish a high-precision regional geomagnetic field model for China, this study integrates Swarm satellite data to investigate the 3D Legendre polynomial model and proposes an enhanced solution method based on singular value decomposition to improve accuracy at higher degrees. Concurrently, the optimal truncation degree of the Legendre polynomial model for each geomagnetic component is determined using K-fold cross-validation. Comparative experiments with Taylor polynomial models, Laguerre polynomial models, and Chebyshev polynomial models validate the advantages of the Legendre polynomial model in terms of truncation degree, computational speed, modeling accuracy, and boundary effects; with overall fitting errors for each component as low as 0.055 nT and boundary errors reaching a minimum of 0.074 nT. Further comparisons with other regional geomagnetic field models and WMM2020 calculation results confirm both the effectiveness and precision advantages of the proposed method along with its corresponding regional geomagnetic field model.

Key words: regional geomagnetic field model, Legendre polynomials, singular value decomposition, K-fold cross-validation, Swarm satellite, WMM2020

CLC Number:

P318

Bo ZHU, Houpu LI, Libo ZHU, Shaofeng BIAN, Cheng CHEN. A regional geomagnetic field model for China based on Swarm satellite data and 3D Legendre polynomials[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(3): 461-472.

Figures/Tables 13

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Tab. 1

Fig. 7

Tab. 2

Fig. 8

Tab. 3

Fig. 9

Tab. 4

References 34

[1]	ALKEN P, THÉBAULT E, BEGGAN C D, et al. International geomagnetic reference field: the thirteenth generation[J]. Earth, Planets and Space, 2021, 73(1): 49.
[2]	CHULLIAT A, BROWN W, ALKEN P, et al. The US/UK world magnetic model for 2020—2025: technical report[R]. Boulder: National Centers for Environmental Information, NOAA, 2020.
[3]	魏毅博. IGRF13和WMM2020模型与实测航磁地磁场数据的差异分析研究[D]. 西安: 长安大学, 2023.
	WEI Yibo. Analysis of the difference between the IGRF13 and WMM2020 models and the measured aeromagnetic geomagnetic field data[D]. Xi'an: Chang'an University, 2023.
[4]	岳颖, 黄静, 李丛. IGRF和WMM模型与地磁台站观测数据对比分析[J]. 测绘标准化, 2018, 34(1): 31-32.
	YUE Ying, HUANG Jing, LI Cong. Comparative analysis of errors from IGRF and WMM model with observations from geomagnetic station[J]. Standardization of Surveying and Mapping, 2018, 34(1): 31-32.
[5]	赵俊, 田源, 张亚军. 全球综合地磁场模型精度评估及统计特征分析[J]. 测绘与空间地理信息, 2024, 47(1): 20-23,28.
	ZHAO Jun, TIAN Yuan, ZHANG Yajun. Accuracy assessment and statistical characteristics analysis of global synthetic geomagnetic field model[J]. Geomatics & Spatial Information Technology, 2024, 47(1): 20-23,28.
[6]	LEE S M, JUNG J, MYUNG H. Geomagnetic field-based localization with bicubic interpolation for mobile robots[J]. International Journal of Control, Automation and Systems, 2015, 13(4): 967-977.
[7]	RIGLER E J, FIORI R A D, PULKKINEN A A, et al. Interpolating geomagnetic observations[M]. Washington, D.C.: American Geophysical Union (AGU), 2019: 15-41.
[8]	LIU Q, XU J, XIAO X, et al. Research on geomagnetic indoor positioning based on different spatial interpolation methods[J]. International Journal of Frontiers in Engineering Technology, 2022, 4(7): 52-59.
[9]	TORTA J M. Modelling by spherical cap harmonic analysis: a literature review[J]. Surveys in Geophysics, 2020, 41(2): 201-247.
[10]	FENG Yan, JIANG Yong, JIANG Yi, et al. Spherical cap harmonic analysis of regional magnetic anomalies based on CHAMP satellite data[J]. Applied Geophysics, 2016, 13(3): 561-569.
[11]	张攀, 杜劲松, 李厚朴. 中国及邻区高分辨率三维岩石圈磁场修正球冠谐模型(CUG＿CLMFM3Dv1)[J]. 地球物理学报, 2024, 67(5): 1866-1880.
	ZHANG Pan, DU Jinsong, LI Houpu. CUG＿CLMFM3Dv1: a high-resolution revised spherical cap harmonic model of three-dimensional lithospheric magnetic field over China and surroundings[J]. Chinese Journal of Geophysics, 2024, 67(5): 1866-1880.
[12]	李明明, 黄显林, 卢鸿谦, 等. 基于矩谐分析的高精度局部地磁场建模研究[J]. 宇航学报, 2010, 31(7): 1730-1736.
	LI Mingming, HUANG Xianlin, LU Hongqian, et al. Modeling of high accuracy local geomagnetic field base on rectangular harmonic analysis[J]. Journal of Astronautics, 2010, 31(7): 1730-1736.
[13]	赵建虎, 刘辉, 王胜平. 局域海洋地磁场矩谐分析建模方法研究[J]. 大地测量与地球动力学, 2009, 29(2): 82-87.
	ZHAO Jianhu, LIU Hui, WANG Shengping. Study on model building in RHA of local ocean geomagnetic field[J]. Journal of Geodesy and Geodynamics, 2009, 29(2): 82-87.
[14]	区家明, 杜爱民, 徐文耀, 等. 小尺度地磁场勒让德多项式建模方法[J]. 地球物理学报, 2012, 55(8): 2669-2675.
	OU Jiaming, DU Aimin, XU Wenyao, et al. The Legendre polynomials modeling method of small-scale geomagnetic fields[J]. Chinese Journal of Geophysics, 2012, 55(8): 2669-2675.
[15]	赵建虎, 王胜平, 刘辉, 等. 海洋局域地磁正常场勒让德多项式模型的建立[J]. 地球物理学进展, 2008, 23(6): 1802-1808.
	ZHAO Jianhu, WANG Shengping, LIU Hui, et al. Construction of normal geomagnetic filed model with Legendre polynomial model in local marine area[J]. Progress in Geophysics, 2008, 23(6): 1802-1808.
[16]	种洋, 柴洪洲. 基于三维Taylor多项式建立局域海洋地磁场模型[J]. 大地测量与地球动力学, 2016, 36(10): 854-858.
	CHONG Yang, CHAI Hongzhou. Regional marine geomagnetic field model reconstruction based on three-dimensional Taylor polynomial[J]. Journal of Geodesy and Geodynamics, 2016, 36(10): 854-858.
[17]	张素琴, 付长华, 赵旭东. 基于三维Taylor模型的福建及邻近地区地磁模型研究[J]. 地球物理学报, 2016, 59(6): 1948-1956.
	ZHANG Suqin, FU Changhua, ZHAO Xudong. Study of regional geomagnetic model of Fujian and adjacent areas based on 3D Taylor polynomial model[J]. Chinese Journal of Geophysics, 2016, 59(6): 1948-1956.
[18]	蒋勇, 姜乙, 冯彦, 等. 基于CHAMP卫星与三维Taylor多项式模型的区域地磁建模研究[J]. 地球物理学报, 2015, 58(9): 3121-3132.
	JIANG Yong, JIANG Yi, FENG Yan, et al. Regional geomagnetic modelling based on the magnetic data of CHAMP satellite and 3D Taylor polynomial[J]. Chinese Journal of Geophysics, 2015, 58(9): 3121-3132.
[19]	柳士俊, 周小刚, 孙涵, 等. 用于区域地磁场模型计算的三维Taylor多项式方法[J]. 地球物理学进展, 2011, 26(4): 1165-1174.
	LIU Shijun, ZHOU Xiaogang, SUN Han, et al. The three dimension Taylor polynomial method for the calculation of regional geomagnetic field model[J]. Progress in Geophysics, 2011, 26(4): 1165-1174.
[20]	FENG Y, SUN H, JIANG Y. Data fitting and modeling of regional geomagnetic field[J]. Applied Geophysics, 2015, 12(3): 303-316.
[21]	TERAO T, OZAKI K, OGITA T. LU-Cholesky QR algorithms for thin QR decomposition[J]. Parallel Computing, 2020, 92: 102571.
[22]	KRISHNAMOORTHY A, MENON D. Matrix inversion using Cholesky decomposition[C]//Proceedings of 2013 Signal Processing: Algorithms, Architectures, Arrangements, and Applications. Poznan: IEEE, 2013: 70-72.
[23]	MA L, DICKSON K, MCALLISTER J, et al. QR decomposition-based matrix inversion for high performance embedded MIMO receivers[J]. IEEE Transactions on Signal Processing, 2011, 59(4): 1858-1867.
[24]	ALLEN G I, GROSENICK L, TAYLOR J. A generalized least-square matrix decomposition[J]. Journal of the American Statistical Association, 2014, 109(505): 145-159.
[25]	VATANKHAH S, LIU S, RENAUT R A, et al. Improving the use of the randomized singular value decomposition for the inversion of gravity and magnetic data[J]. Geophysics, 2020, 85(5): G93-G107.
[26]	SOLOVYEV SA, TORDEUX S. Large SVD computations for analysis of inverse problems in geophysics[C]//Proceedings of the 11th World Congress on Computational Mechanics. Barcelona: CIMNE, 2014: 2861-2869.
[27]	YANAI H, TAKEUCHI K, TAKANE Y. Singular value decomposition (SVD)[M]//Projection matrices, generalized inverse matrices, and singular value decomposition. New York: Springer, 2011: 125-149.
[28]	郭文彬. 奇异值分解及其在广义逆理论中的应用[D]. 上海: 华东师范大学, 2004.
	GUO Wenbin. Singular value decomposition and its application in the theory of generalized inverses[D]. Shanghai: East China Normal University, 2004.
[29]	聂琳娟, 许智铭. 利用SWARM星群磁测数据反演地球主磁场模型WHU-MM01[J]. 测绘地理信息, 2020, 45(5): 16-19.
	NIE Linjuan, XU Zhiming. The inversion of the earth's main magnetic field model WHU-MM01 with SWARM satellite magnetic data[J]. Journal of Geomatics, 2020, 45(5): 16-19.
[30]	刘长青. 基于Swarm卫星和地面地磁台站数据的区域地磁场建模研究[D]. 郑州: 信息工程大学, 2023.
	LIU Changqing. Research on regional geomagnetic field modeling based on Swarm satellite and ground geomagnetic station data[D]. Zhengzhou: Information Engineering University, 2023.
[31]	刘爽. 基于Swarm卫星数据的卫星高度区域地磁场模拟研究[D]. 南京: 南京信息工程大学, 2022.
	LIU Shuang. Modelling of the regional geomagnetic field at satellite altitudes based on Swarm satellite data[D]. Nanjing: Nanjing University of Information Science and Technology, 2022.
[32]	JUNG K, BAE D H, UM M J, et al. Evaluation of nitrate load estimations using neural networks and canonical correlation analysis with K-fold cross-validation[J]. Sustainability, 2020, 12(1): 400.
[33]	NTI I K, BOATENG O N, ANING J. Performance of machine learning algorithms with different K values in K-fold cross-validation[J]. International Journal of Information Technology and Computer Science (IJITCS), 2021, 13(6): 61-71.
[34]	MARCOT B G, HANEA A M. What is an optimal value of K in K-fold cross-validation in discrete Bayesian network analysis?[J]. Computational Statistics, 2021, 36(3): 2009-2031.

多项式模型	地磁分量	最佳截止阶数	平均拟合误差/nT	平均预测误差/nT
泰勒	总强度F	4	41.515	41.597
	X分量	4	23.980	24.010
	Y分量	4	30.560	30.609
	Z分量	4	41.346	41.407
拉盖尔	总强度F	17	0.379	0.384
	X分量	15	0.603	0.611
	Y分量	15	0.942	0.954
	Z分量	15	2.978	2.989
切比雪夫	总强度F	22	0.112	0.117
	X分量	18	0.304	0.313
	Y分量	17	0.561	0.575
	Z分量	13	2.946	2.963
勒让德	总强度F	26	0.052	0.073
	X分量	23	0.209	0.233
	Y分量	23	0.451	0.489
	Z分量	15	2.929	2.961

模型	总场强F	X分量	Y分量	Z分量
泰勒多项式模型	41.525	23.989	30.562	41.392
拉盖尔多项式模型	0.321	0.743	0.868	3.023
切比雪夫多项式模型	0.111	0.289	0.555	2.946
勒让德多项式模型	0.055	0.207	0.452	2.934

多项式模型	总场强F	X分量	Y分量	Z分量
泰勒	113.885	50.643	41.073	97.788
拉盖尔	0.331	1.208	1.143	2.857
切比雪夫	0.213	0.228	0.386	0.480
勒让德	0.074	0.131	0.360	0.377

模型	总场强F	X分量	Y分量	Z分量
文献[30]泰勒多项式模型	20.28	22.38	14.52	32.06
文献[31]泰勒多项式模型	54.1	19.1	48.8	67.9
本文泰勒多项式模型	41.525	23.989	30.562	41.392
本文勒让德多项式模型	0.055	0.207	0.452	2.934
WMM2020	46.581	161.924	45.932	133.256