Abstract
This study explores machine learning models to gain insights into dynamics of ionospheric irregularities over geodetic receivers in Mbarara (0.60° S, 30.74° E) and Kigali (1.94° S, 30.09° E). A seven-year rate of total electron content index (ROTI) database and two modeling approaches (multivariate and univariate) were employed. The motivation was to treat the database with time series techniques following a case study with and without the influence of solar wind parameters. The objective is to examine how each approach reconstructs the morphology of ROTI within 3-h time steps over a 24-h cycle. To achieve this, five machine learning models, including extreme gradient boosting (XGBoost), random forest (RF), bidirectional long-short term memory (BLSTM), unidirectional long-short term memory (LSTM) and nonlinear autoregressive with eXogenous input (NARX), were developed and evaluated. Test results demonstrate significant performance variations highlighting comparable ROTI reconstructions in the absence of the solar wind features. The RF model exhibited superior performance with the lowest mean absolute errors of 0.03 and 0.07 TECU/min and accuracies of 93% and 75% under multivariate and univariate modeling, respectively. Based on the RF model’s performance, we employed an extended database over the Ugandan (Mbar) station for further model development and validated its efficiency over a station in Rwanda (Nurk). The results provided promising insights, emphasizing the need for future research dedicated to robust and enhanced nowcasting models that leverage long-term ionospheric data, especially in regions with limited scintillation monitors.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
The IGS data over the African region used in this work can be found in the GNSS archives of the University of California San Diego, available at http://garner.ucsd.edu/pub/rinex/. The satellite navigation files can be retrieved from NOAA’s National Geodetic Survey Continuously Operating Reference Station (CORS) database at https://geodesy.noaa.gov/corsdata/rinex/. Finally, the geomagnetic activity indices and TEC analysis software can be accessed from https://wdc.kugi.kyoto-u.ac.jp/, https://omniweb.gsfc.nasa.gov/ and https://seemala.blogspot.com/, respectively.
References
Aarons J, Mendillo M, Yantosca R (1996) GPS phase fluctuations in the equatorial region during sunspot minimum. J Geophys Res Sp Phys 101(A12):26851–26862. https://doi.org/10.1029/96JA00981
Abadi P, Saito S, Srigutomo W (2014) Low-latitude scintillation occurrences around the equatorial anomaly crest over Indonesia. Ann Geophys 32(1):7–17. https://doi.org/10.5194/angeo-32-7-2014
Akala AO, Amaeshi LLN, Doherty PH, Groves KM, Carrano CS, Bridgwood CT, Seemala GK, Somoye EO (2014) Characterization of GNSS scintillations over Lagos, Nigeria during the minimum and ascending phases (2009–2011) of solar cycle 24. Adv Sp Res 53(1):37–47. https://doi.org/10.1016/j.asr.2013.09.034
Amaechi PO, Oyeyemi EO, Akala AO, Amory-Mazaudier C (2020) Geomagnetic activity control of irregularities occurrences over the crests of the African EIA. Earth Sp Sci 7(7):e2020EA001183. https://doi.org/10.1029/2020EA001183
Arif Ali Z, Abduljabbar Z, Tahir H, Bibo Sallow A, Almufti SM (2023) eXtreme gradient boosting algorithm with machine learning: a review. Acad J Nawroz Univ 12(2):320–334. https://doi.org/10.25007/ajnu.v12n2a1612
Atabati A, Alizadeh M, Schuh H, Tsai L-C (2021) Ionospheric scintillation prediction on S4 and ROTI parameters using artificial neural network and genetic algorithm. Remote Sens 13(11):2092. https://doi.org/10.3390/rs13112092
Basu S, Groves KM, Quinn JM, Doherty P (1999) A comparison of TEC fluctuations and scintillations at Ascension Island. J Atmos Solar-Terrestrial Phys 61(16):1219–1226. https://doi.org/10.1016/S1364-6826(99)00052-8
Bhattacharyya A (2022) Equatorial plasma bubbles: a review. Atmosphere (basel) 13(10):1637. https://doi.org/10.3390/atmos13101637
Bolaji OS, Adebiyi SJ, Fashae JB (2019) Characterization of ionospheric irregularities at different longitudes during quiet and disturbed geomagnetic conditions. J Atmos Solar-Terrestrial Phys 182:93–100. https://doi.org/10.1016/j.jastp.2018.11.007
Carrano CS, Groves KM, Rino CL (2019) On the relationship between the rate of change of total electron content index (ROTI), irregularity strength (CkL), and the scintillation index (S4). J Geophys Res Sp Phys 124(3):2099–2112. https://doi.org/10.1029/2018JA026353
Carter BA, Retterer JM, Yizengaw E, Wiens K, Wing S, Groves K, Caton R, Bridgwood C, Francis M, Terkildsen M, Norman R, Zhang K (2014) Using solar wind data to predict daily GPS scintillation occurrence in the African and Asian low-latitude regions. Geophys Res Lett 41(23):8176–8184. https://doi.org/10.1002/2014GL062203
Carter BA, Currie JL, Dao T, Yizengaw E, Retterer J, Terkildsen M, Groves K, Caton R (2020) On the assessment of daily equatorial plasma bubble occurrence modeling and forecasting. Sp Weather 18(9):e2020SW002555. https://doi.org/10.1029/2020SW002555
Carvalho O, da Silvade Almeida Nava Alves PAA, de La Cruz Cueva RY, Filho AOB (2022) Nowcasting of amplitude ionospheric scintillation based on machine learning techniques. IEEE Trans Aerosp Electron Syst 58(6):4917–4927. https://doi.org/10.1109/TAES.2022.3188741
Chandan K, Gopi KS, Dadaso JS, Acharya R (2021) Reckoning ionospheric scintillation S4 from ROTI over Indian region. Adv Sp Res 69(2):915–925. https://doi.org/10.1016/j.asr.2021.10.026
Cherniak I, Krankowski A, Zakharenkova I (2014) Observation of the ionospheric irregularities over the Northern Hemisphere: methodology and service. Radio Sci 49(8):653–662. https://doi.org/10.1002/2014RS005433
Correia E, Tadeu de Assis Honorato Muella M, Alfonsi L, dos Santos Prol F, de Oliveira Camargo P (2019) GPS scintillations and total electron content climatology in the Southern American Sector. In: Accuracy of GNSS methods. IntechOpen, pp 47–70
De Franceschi G, Spogli L, Alfonsi L, Romano V, Cesaroni C, Hunstad I (2019) The ionospheric irregularities climatology over Svalbard from solar cycle 23. Sci Rep 9(1):9232. https://doi.org/10.1038/s41598-019-44829-5
de Lima GRT, Stephany S, de Paula ER, Batista IS, Abdu MA (2015) Prediction of the level of ionospheric scintillation at equatorial latitudes in Brazil using a neural network. Sp Weather 13(8):446–457. https://doi.org/10.1002/2015SW001182
de Oliveira MA, Costa E, de Paula ER, Perrella WJ, Monico JFG (2014) Extended ionospheric amplitude scintillation model for GPS receivers. Radio Sci 49(5):315–329. https://doi.org/10.1002/2013RS005307
DiPietro R, Hager GD (2020) Deep learning: RNNs and LSTM. In: Handbook of medical image computing and computer assisted intervention. Elsevier, pp 503–519
Dugassa T, Bosco Habarulema J, Nigussie M (2019) Investigation of the relationship between the spatial gradient of total electron content (TEC) between two nearby stations and the occurrence of ionospheric irregularities. Ann Geophys 37(6):1161–1180. https://doi.org/10.5194/angeo-37-1161-2019
Haji SH, Abdulazeez AM (2021) Comparison of optimization techniques based on gradient descent algorithm: a review. PalArch’s J Archaeol Egypt/egyptology 18(4):2715–2743
Jiao Y, Morton YT, Taylor S, Pelgrum W (2013) Characterization of high-latitude ionospheric scintillation of GPS signals. Radio Sci 48(6):698–708. https://doi.org/10.1002/2013RS005259
Kingma D, Ba J (2015) Adam: a method for stochastic optimization. In: Proceedings of ICLR-2015. https://doi.org/10.48550/arXiv.1412.6980
Kogogin DA, Nasyrov IA, Maksimov DS, Zagretdinov RV (2021) Using Android smartphones with dual-frequency multi-GNSS receiver to measure the total electron content of the ionosphere. J Phys Conf Ser 1:012025. https://doi.org/10.1088/1742-6596/1991/1/012025
Kumar V, Singh Aydav PS, Minz S (2022) Multi-view ensemble learning using multi-objective particle swarm optimization for high dimensional data classification. J King Saud Univ Comput Inf Sci 34(10):8523–8537. https://doi.org/10.1016/j.jksuci.2021.08.029
Lamb K, Malhotra G, Vlontzos A, Wagstaff E, Baydin AG, Bhiwandiwalla A, Gal Y, Kalaitzis A, Reina A, Bhatt A (2019) Prediction of GNSS phase scintillations: a machine learning approach. arXiv Prepr. https://doi.org/10.48550/arXiv.1910.01570
Li W, Song S, Jin X (2022) Ionospheric scintillation monitoring with ROTI from geodetic receiver: limitations and performance evaluation. Radio Sci 57(5):1–15. https://doi.org/10.1029/2021RS007420
Liu L, Morton YJ, Liu Y (2021) Machine learning prediction of storm-time high-latitude ionospheric irregularities from GNSS-derived ROTI maps. Geophys Res Lett 48:20. https://doi.org/10.1029/2021GL095561
Montesinos López OA, Montesinos López A, Crossa J (2022) Multivariate statistical machine learning methods for genomic prediction. Springer, Cham
Mungufeni P, Habarulema JB, Jurua E (2016) Trends of ionospheric irregularities over African low latitude region during quiet geomagnetic conditions. J Atmos Solar-Terrestrial Phys 138:261–267. https://doi.org/10.1016/j.jastp.2016.01.015
Nguyen CT, Oluwadare ST, Le NT, Alizadeh M, Wickert J, Schuh H (2021) Spatial and temporal distributions of ionospheric irregularities derived from regional and global ROTI maps. Remote Sens 14(1):10. https://doi.org/10.3390/rs14010010
Okoh D, Rabiu B, Shiokawa K, Otsuka Y, Segun B, Falayi E, Onwuneme S, Kaka R (2017) First study on the occurrence frequency of equatorial plasma bubbles over West Africa using an all-sky airglow imager and GNSS receivers. J Geophys Res Sp Phys 122(12):12430–12444. https://doi.org/10.1002/2017JA024602
Olwendo J, Cilliers PJ, Ming O (2021) Monthly trends in temporal and spatial distribution of Ionospheric Irregularities across the African region during the descending phase of solar cycle 24. Adv Sp Res 67(10):3187–3201. https://doi.org/10.1016/j.asr.2021.01.052
Otsuka Y, Aramaki T, Ogawa T, Saito A (2006) A statistical study of ionospheric irregularities observed with a GPS network in Japan. Geophys Monogr Ser 167:271–281. https://doi.org/10.1029/167GM21
Paznukhov VV, Carrano CS, Doherty PH, Groves KM, Caton RG, Valladares CE, Seemala GK, Bridgwood CT, Adeniyi J, Amaeshi LLN, Damtie B, D’Ujanga Mutonyi F, Ndeda JOH, Baki P, Obrou OK, Okere B, Tsidu GM (2012) Equatorial plasma bubbles and L-band scintillations in Africa during solar minimum. Ann Geophys 30(4):675–682. https://doi.org/10.5194/angeo-30-675-2012
Pi X, Mannucci AJ, Lindqwister UJ, Hod UCM (1997) Monitoring of global ionospheric irregularities using the worldwide GPS network. Geophys Res Lett 24(18):2283–2286. https://doi.org/10.1029/97GL0227
Priyadarshi S (2015) A review of ionospheric scintillation models. Surv Geophys 36(1):295–324. https://doi.org/10.1007/s10712-015-9319-1
Reddy SA, Forsyth C, Aruliah A, Smith A, Bortnik J, Aa E, Kataria DO, Lewis G (2023) Predicting swarm equatorial plasma bubbles via machine learning and shapley values. J Geophys Res Sp Phys 128:6. https://doi.org/10.1029/2022JA031183
Schneider P, Xhafa F (2022) Anomaly detection. In: Anomaly detection and complex event processing over IoT data streams. Elsevier, pp 49–66
Shao Y, Zhao J, Xu J, Fu A, Li M (2022) Application of rainfall-runoff simulation based on the NARX dynamic neural network model. Water 14(13):2082. https://doi.org/10.3390/w14132082
Siami-Namini S, Tavakoli N, Namin AS (2019) The performance of LSTM and BiLSTM in forecasting time series. In: 2019 IEEE international conference on big data (big data). IEEE, pp 3285–3292
Singh RP, Patel RP, Singh AK (2004) Effect of solar and magnetic activity on VHF scintillations near the equatorial anomaly crest. Ann Geophys 22(8):2849–2860. https://doi.org/10.5194/angeo-22-2849-2004
Sripathi S, Sreekumar S, Banola S (2018) Characteristics of equatorial and low-latitude plasma irregularities as investigated using a meridional chain of radio experiments over India. J Geophys Res Sp Phys 123:4364–4380
Sun L, Schulz K (2015) The improvement of land cover classification by thermal remote sensing. Remote Sens 7(7):8368–8390. https://doi.org/10.3390/rs70708368
Taabu SD, D’ujanga FM, Ssenyonga T (2016) Prediction of ionospheric scintillation using neural network over East African region during ascending phase of sunspot cycle 24. Adv Sp Res 57(7):1570–1584. https://doi.org/10.1016/j.asr.2016.01.014
Tete S, Otsuka Y, Zahra WK, Mahrous A (2023) Machine learning approach for ionospheric scintillation prediction on ROTI parameter over the African region during solar cycle 24. Adv Sp Res. https://doi.org/10.1016/j.asr.2023.12.026
Veettil SV, Haralambous H, Aquino M (2017) Observations of quiet-time moderate midlatitude L-band scintillation in association with plasma bubbles. GPS Solut 21:113–124
Xu F, Zishen L, Zhang K, Ningbo W, Wu S, Andong H, Holden L (2020) An investigation of optimal machine learning methods for the prediction of ROTI. J Geod Geoinf 3(2):1–15. https://doi.org/10.11947/j.JGGS.2020.0201
Yang Z, Liu Z (2016) Correlation between ROTI and Ionospheric scintillation indices using Hong Kong low-latitude GPS data. GPS Solut 20(4):815–824. https://doi.org/10.1007/s10291-015-0492-y
Yizengaw E, Groves K (2020) Forcing from lower thermosphere and quiet time scintillation longitudinal dependence. Sp Weather 18(11):e2020SW002610. https://doi.org/10.1029/2020SW002610
Acknowledgements
The authors of this work acknowledge the University of California San Diego, International GNSS services and SONEL for making the geodetic receiver data available over the African region. We also thank the NOAA’s National Geodetic Survey Continuously Operating Reference Station (CORS) for providing the navigation files. Finally, we acknowledge the World Data Centre for Geomagnetism, Kyoto, Japan and the Goddard Space Flight Centre, NASA, for making geomagnetic activity indices accessible. Y.O. is supported by a JSPS KAKENHI Grant (20H00197, 21H04518, 22K21345), JSPS Bilateral Joint Research Projects (JPJSBP120226504), and JSPS Core-to-Core Program, B. Asia-Africa Science Platforms.
Funding
No funds, grants, or other support were received to conduct this study.
Author information
Authors and Affiliations
Contributions
Stephen Tete processed the data, developed the models and wrote the manuscript. Yuichi Otsuka led the collaborative manuscript edits and provided advice and paper review. Waheed K. Zahra provided manuscript edits and advice. Ayman Mahrous guided the experimental design and provided supervision and manuscript review. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Conflict of interest
The authors declare no conflicts of interest related to this work.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Tete, S., Otsuka, Y., Zahra, W.K. et al. Leveraging machine learning techniques and GPS measurements for precise TEC rate predictions. GPS Solut 28, 115 (2024). https://doi.org/10.1007/s10291-024-01652-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10291-024-01652-4