The Improved Two‐Dimensional Artificial Neural Network‐Based Ionospheric Model (ANNIM)

Ram, S. Tulasi; Gowtam, V. Sai; Mitra, Arka; Reinisch, B. W.

doi:10.1029/2018ja025559

Cited by 41 publications

(36 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method used to reduce the discrepancies between multiple instrument measurements shows a strong ability to improve the ionospheric model's accuracy. In previous studies, many researchers focused on ionospheric modeling based on deep learning with GNSS space-borne occultations or ground-based observations [15,17,24]. In the literature [15], the measurements obtained from 59 ionosondes were used to develop the global foF2 model by a feed-forward neural network.…”

Section: Discussionmentioning

confidence: 99%

“…In addition, due to the limitation of geographic distribution of the selected stations, this model fails to capture the ionospheric peak parameters information over the ocean exactly. In order to solve the problem, a previous study [17] multiplied the instrument's measurements to build the improved ionospheric model ANNIM, such as COSMIC, CHAMP, GRACE and ionosonde. This model was shown represent global ionospheric spatiotemporal characteristics well, but the RMSEs in the solar maximum year (2002) and solar minimum year (2009) for NmF2 were 3.4 × 105 electrons/cm 3 and 1.5 × 105 electrons/cm 3 , and the corresponding RMSEs for hmF2 were 29 km and 25 km, respectively.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Advanced Machine Learning Optimized by The Genetic Algorithm in Ionospheric Models Using Long-Term Multi-Instrument Observations

Zhao

Hé

et al. 2020

Remote Sensing

View full text Add to dashboard Cite

The ionospheric delay is of paramount importance to radio communication, satellite navigation and positioning. It is necessary to predict high-accuracy ionospheric peak parameters for single frequency receivers. In this study, the state-of-the-art artificial neural network (ANN) technique optimized by the genetic algorithm is used to develop global ionospheric models for predicting foF2 and hmF2. The models are based on long-term multiple measurements including ionospheric peak frequency model (GIPFM) and global ionospheric peak height model (GIPHM). Predictions of the GIPFM and GIPHM are compared with the International Reference Ionosphere (IRI) model in 2009 and 2013 respectively. This comparison shows that the root-mean-square errors (RMSEs) of GIPFM are 0.82 MHz and 0.71 MHz in 2013 and 2009, respectively. This result is about 20%-35% lower than that of IRI. Additionally, the corresponding hmF2 median errors of GIPHM are 20% to 30% smaller than that of IRI. Furthermore, the ANN models present a good capability to capture the global or regional ionospheric spatial-temporal characteristics, e.g., the equatorial ionization anomaly and Weddell Sea anomaly. The study shows that the ANN-based model has a better agreement to reference value than the IRI model, not only along the Greenwich meridian, but also on a global scale. The approach proposed in this study has the potential to be a new three-dimensional electron density model combined with the inclusion of the upcoming Constellation Observing System for Meteorology, Ionosphere and Climate (COSMIC-2) data.Remote Sens. 2020, 12, 866 2 of 17 space-borne and ground-based instruments and data processed techniques, more attention has been paid to the development and improvement of these ionospheric models [3].In the previous studies, the artificial neural network (ANN), nonlinear least squares and AdaBoost techniques have been used to predict ionospheric variations with a satisfactory accuracy [4][5][6][7]. Among them, the ANN technique has proven to be a successful tool in the modeling of ionospheric variations as well as solving the forecast problems in many geophysical applications over a single station, regional area and global scale [3,[8][9][10]. The neural network was first applied to build the foF2 model over the Grahamstown station (33 • S, 26 • E) using the data range set of 1973-1983 [11]. This model can predict the daily noon foF2 value with a root-mean-square error (RMSE) of 0.95 MHz and the monthly averaged RMSE value was 0.48 MHz. Afterward, this approach has been widely applied in the Asia/Pacific sector [5], equatorial region [12], European region [13] and the auroral zone [14]. These studies showed an improvement in the accuracy of the ANN-based models over the International Reference Ionosphere (IRI) model.With the advancement of data acquisition, more ionospheric measurements have become available for the global modelling of foF2. For instance, Oyeyemi et al. [15] used the foF2 data from 59 globally distributed ionospheric stations to establish th...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Advanced Machine Learning Optimized by The Genetic Algorithm in Ionospheric Models Using Long-Term Multi-Instrument Observations

Zhao

Hé

et al. 2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Recently, Sai and Tulasi [24] developed ANNIM to predict global NmF2 based on COSMIC data. Afterwards, Tulasi et al [25] presented an improved version of ANNIM by assimilating more IRO measurements as well as ionosonde data using a modified spatial gridding approach based on the magnetic dip latitudes. They used an Artificial Neural Networks technique, which enabled the estimation of the coefficients without knowing the base function of independent variables in auroral region as a result of the complex interplay between different geophysical processes [63].…”

Section: Discussionmentioning

confidence: 99%

A New Empirical Model of NmF2 Based on CHAMP, GRACE, and COSMIC Radio Occultation

et al. 2019

View full text Add to dashboard Cite

To facilitate F2-layer peak density (NmF2) modeling, a nonlinear polynomial model approach based on global NmF2 observational data from ionospheric radio occultation (IRO) measurements onboard the CHAMP, GRACE, and COSMIC satellites, is presented in this paper. We divided the globe into 63 slices from 80°S to 80°N according to geomagnetic latitude. A Nonlinear Polynomial Peak Density Model (NPPDM) was constructed by a multivariable least squares fitting to NmF2 measurements in each latitude slice and the dependencies of NmF2 on solar activity, geographical longitude, universal time, and day of year were described. The model was designed for quiet and moderate geomagnetic conditions (Ap ≤ 32). Using independent radio occultation data, quantitative analysis was made. The correlation coefficients between NPPDM predictions and IRO data were 0.91 in 2002 and 0.82 in 2005. The results show that NPPDM performs better than IRI2016 and Neustrelitz Peak Density Model (NPDM) under low solar activity, while it undergoes performance degradation under high solar activity. Using data from twelve ionosonde stations, the accuracy of NPPDM was found to be better than that of NPDM and comparable to that of IRI2016. Additionally, NPPDM can well simulate the variations and distributions of NmF2 and describe some ionospheric features, including the equatorial ionization anomaly, the mid-latitude trough, and the wavenumber-four longitudinal structure.

show abstract

“…Recently, a global ANNIM-2D was developed to predict N m F 2 and h m F 2 by implementing the machine learning technique (Sai Gowtam & Tulasi Ram, 2017a;Tulasi Ram et al, 2018). ANNIM-2D is essentially an empirical model developed by training the Artificial Neural Networks (ANNs) with long-term data of N m F 2 and h m F 2 from ground-based global Digisonde observations, and the GPS-RO missions.…”

Section: Model Description and Methodologymentioning

confidence: 99%

“…It is clear that ANNIM-2D could reproduce the EHA phenomena, as the model was heavily waited on COSMIC data ( Figure 1). Further, Tulasi Ram et al (2018) reported that the ANNIM-2D has an excellent ability to predict the N m F 2 and h m F 2 under disturbed space weather conditions. Hence, the model results from ANNIM-2D are utilized in the present study to explore the equatorial and low-latitude variation of EHA during the main phase of St. Patrick's Day geomagnetic storm.…”

Section: Model Description and Methodologymentioning

confidence: 99%

Variation of the Equatorial Height Anomaly During the Main Phase of 2015 St. Patrick's Day Geomagnetic Storm Using ANNIM and TIEGCM

et al. 2019

Self Cite

View full text Add to dashboard Cite

A double‐peak structure in the peak height of ionospheric F2 layer around ±10o geomagnetic latitudes similar to the equatorial ionization anomaly was recently reported. This unique feature was referred as the equatorial height anomaly (EHA). In the present paper, a simulation study is carried out using the data‐driven artificial neural network‐based two‐dimensional ionospheric model (ANNIM‐2D) and the physics‐based thermosphere‐ionosphere‐electrodynamics general circulation model (TIEGCM) to understand the local time and latitudinal variation of EHA during the main phase of St. Patrick's Day geomagnetic storm. Both the ANNIM‐2D and TIEGCM consistently show pronounced EHA during the main phase of the geomagnetic storm. Further, the local time of EHA development on the storm day is much earlier (nearly 2 hr) than the quiet time over Brazilian sector (90°W). The TIEGCM simulation revealed that the storm time enhancement of the equatorial fountain associated with the enhanced equatorial zonal electric field is the main controlling factor for the pronounced EHA during the main phase. The storm time meridional neutral winds positively contribute to the development of EHA. This study revealed the direct manifestation of the storm time‐enhanced plasma fountain on the EHA.

show abstract

The Improved Two‐Dimensional Artificial Neural Network‐Based Ionospheric Model (ANNIM)

Cited by 41 publications

References 69 publications

Advanced Machine Learning Optimized by The Genetic Algorithm in Ionospheric Models Using Long-Term Multi-Instrument Observations

Advanced Machine Learning Optimized by The Genetic Algorithm in Ionospheric Models Using Long-Term Multi-Instrument Observations

A New Empirical Model of NmF2 Based on CHAMP, GRACE, and COSMIC Radio Occultation

Variation of the Equatorial Height Anomaly During the Main Phase of 2015 St. Patrick's Day Geomagnetic Storm Using ANNIM and TIEGCM

Contact Info

Product

Resources

About