Influence of DE3 tide on the equinoctial asymmetry of the zonal mean ionospheric electron density
© Ren et al.; licensee Springer. 2014
Received: 29 March 2014
Accepted: 4 September 2014
Published: 17 September 2014
Through respectively adding September DE3 tide and March DE3 tide at the low boundary of Global Coupled Ionosphere-Thermosphere-Electrodynamics Model, Institute of Geology and Geophysics, Chinese Academy of Sciences (GCITEM-IGGCAS), we simulate the influence of DE3 tide on the equinoctial asymmetry of the zonal mean ionospheric electron density. The influence of DE3 tide on the equinoctial asymmetry of the zonal mean electron density varies with latitude, altitude, and solar activity level. Compared with the density driven by the September DE3 tide, the March DE3 tide mainly decreases the lower ionospheric zonal mean electron density and mainly increases the electron density at higher ionosphere. In the low-latitude ionosphere, DE3 tide drives an equatorial ionization anomaly (EIA) structure at higher ionosphere in the relative difference of zonal mean electron density, which suggests that DE3 tide affects the longitudinal mean equatorial vertical E × B plasma drifts. Although the lower ionospheric equinoctial asymmetry driven by DE3 tide mainly decreases with the increase of solar activity, the asymmetry at higher ionosphere mainly increases with solar activity. However, EIA in equinoctial asymmetry mainly decreases with the increase of solar activity.
KeywordsIonospheric equinoctial asymmetry DE3 tide Ionosphere–atmosphere coupling
The ionospheric seasonal variability is an unresolved important question in ionospheric research. The previous works are mainly focused on the differences between summer and winter and between solstice and equinox (e.g., Zhao et al. 2007). However, since Titheridge (1973) first studied the difference between the mid-latitude ionospheric electron density in March equinox and that in September equinox, many researchers paid attention to the ionospheric equinoctial asymmetry (e.g., Liu et al. 2010). Now, the ionospheric equinoctial asymmetry, with some ionospheric parameters being larger during one equinox than those during the other one, is an important feature of ionospheric seasonal variation, and the equinoctial asymmetry had been found in a series of ionospheric parameters, such as total electron content (TEC), topside ionospheric electron density, ionospheric electron density profiles, the F layer height, electron temperatures, ion temperatures, field-parallel drifts, and field-perpendicular plasma drifts (e.g., Titheridge and Buonsanto 1983; Aruliah et al. 1996; Balan et al. 1998, 2006; Bailey et al. 2000; Kawamura et al. 2002; Unnikrishnan et al. 2002; Zhang et al. 2004; Zhao et al. 2007; Wan et al. 2008; Ren et al. 2011a).
Although the earlier researchers are mainly focused on the ionospheric equinoctial asymmetry in the high- and mid-latitude, recent research suggested that the low-latitude ionospheric equinoctial asymmetry is also important. Liu et al. (2010) investigated the daytime ionospheric behaviors around equinoxes and found that the equinoctial asymmetry in the ionospheric plasma density during low solar activity is mainly a low-latitude phenomenon. Chen et al. (2012) studied the equinoctial asymmetry in solar activity variations of NmF2 and TEC. With vertical drift (V⊥) data observed by ROCSAT-1, Ren et al. (2011a) studied the equinoctial asymmetry of equatorial V⊥ and suggested that such asymmetry can partly explain the equinoctial asymmetry in daytime low-latitude ionospheric plasma density observed by Liu et al. (2010). Based on the average values of ionosonde hmF2 data acquired from an African equatorial station, Adebesin et al. (2013) also found the equinoctial asymmetry in equatorial V⊥. Using multi-instrument observations, Sripathi et al. (2011) studied equinoctial asymmetry in the equatorial spread F (ESF) irregularities over Indian region. Based on the observations using FORMOSAT-3/COSMIC GPS RO technique during a solar minimum year, Brahmanandam et al. (2012) observed the equinoctial asymmetry in the global S4 index. They both suggested that the equinoctial asymmetry in equatorial V⊥ plays an important role in the generation of these asymmetries. Ren et al. (2012a) simulated the influence of atmospheric tides on equinoctial asymmetry of V⊥ with TIDM-IGGCAS-II ionospheric dynamo model (Ren et al. 2008) and found that the simulated equinoctial asymmetry in V⊥ are mainly driven by the migrating diurnal tide, the migrating semidiurnal tide, the eastward propagating non-migrating diurnal tide with zonal wave number 2 (DE2), and the eastward propagating non-migrating diurnal tide with zonal wave number 3 (DE3).
DE3 is excited in the tropical troposphere, propagates vertically, and is sometimes the single largest tidal component above the mesopause (Forbes et al. 2008; Oberheide and Forbes 2008; Mukhtarov et al. 2009; Pancheva and Mukhtarov 2010). There are growing observational evidences that DE3 is one of the main sources of the longitudinal wavenumber-4 structure (WN4) in the ionosphere. Although some observational and numerical simulation evidences indicate that not only DE3 tide but also SPW4 and SE2 waves have contribution to the ionospheric WN4, DE3 still attracts a considerable attention. Based on the satellite observations, Oberheide and Forbes (2008), Ren et al. (2009a, 2010), and Wan et al. (2010) all found that the zonal wind of DE3 is stronger and occurs primarily during northern summer and autumn and DE3 tide shows obvious equinoctial asymmetry. In the previous research, many researchers simulated the influence of DE3 tide on ionosphere (e.g., Ren et al. 2011b, 2012b; Wan et al. 2012; Wu et al. 2012). However, most of these work mainly paid attention to the coupling between DE3 and the ionospheric WN4. For example, Jin et al. (2008) and Ren et al. (2010) mainly simulated the influence of DE3 tide on the WN4 structure in equatorial vertical E × B plasma drifts.
DE3 tide not only drives the ionospheric WN4 structures but also affects the mean states of ionosphere and thermosphere. Forbes et al. (1993) utilized NCAR TIGCM to simulate tidal influence on ionosphere and thermosphere and found that the upward propagating migrating tides can accelerate, heat, and mix composition in the coupled ionosphere-thermosphere system. Jones et al. (2014) simulated the impacts of vertically propagating tides on the mean state of the ionosphere and thermosphere and found that the non-migrating tide DE3 also can affect the ionospheric mean state. Because the intra-annual variation of DE3 tide shows equinoctial asymmetry, DE3 tide may also affect the equinoctial asymmetry of the ionospheric mean states. Not only DE3 tide but also the migrating diurnal and semidiurnal tides show obvious equinoctial asymmetry and may affect the ionospheric equinoctial asymmetry (see Forbes et al. 1993, Mukhtarov et al. 2009, and Jones et al. 2014). However, because of the nonlinearly interaction and nonlinearly interaction between different tides, the tides' influences on ionospheric equinoctial asymmetry are very complex. Hence, in this paper, we will mainly pay attention to the influence of DE3 tide and simulate the influence of DE3 tide on the equinoctial asymmetry of the ionospheric zonal mean states with Global Coupled Ionosphere-Thermosphere-Electrodynamics Model, Institute of Geology and Geophysics, Chinese Academy of Sciences (GCITEM-IGGCAS) and Three-Dimensional Theoretical Ionospheric Model of the Earth, Institute of Geology and Geophysics, Chinese Academy of Sciences (TIME3D-IGGCAS).
Model descriptions and inputs
To simulate the complex and highly coupled physical and chemical processes in the ionosphere-thermosphere system, we had developed a global coupled ionosphere-thermosphere model (GCITEM-IGGCAS). GCITEM-IGGCAS is a three-dimensional (3-D) code with 5° latitude by 7.5° longitude cells in a spherical geographical coordinate system, which bases on an altitude grid. This model self-consistently calculates the time-dependent 3-D structures of the main thermospheric and ionospheric parameters in the height range from 90 to 600 km, including the neutral number density of major species O2, N2, and O and minor species N(2D), N(4S), NO, He, and H; ion number densities of O+, O2+, N2+, NO+, N+, and electrons; neutral, electron, and ion temperatures; neutral wind vectors; and ionospheric electric field. GCITEM-IGGCAS can reproduce the main features of the thermosphere and ionosphere. The details of GCITEM-IGGCAS are given in Ren et al. (2009b).
Because it uses a spherical geographical coordinate system, GCITEM-IGGCAS could not self-consistently calculate the heat flux and plasma flux at the upper boundary and had to obtain these fluxes from empirical models. Theoretical ionospheric model based on the closed geomagnetic tubes can self-consistently calculate these fluxes. TIME3D-IGGCAS is a three-dimensional theoretical ionospheric model in realistic geomagnetic fields (Ren et al. 2012c). This model covers the whole ionosphere and whole plasmasphere and can self-consistently calculate the time-dependent three-dimensional structures of the main ionospheric and plasmaspheric parameters in realistic geomagnetic fields, including ion number densities of O+, H+, He+, NO+, O2+, N2+, and electron; electron and ion temperature; and ion velocity vectors. TIME3D-IGGCAS and GCITEM-IGGCAS can two-way couple with each other. In this condition, GCITEM-IGGCAS will close its ionosphere module and run TIME3D-IGGCAS as its ionosphere-plasmasphere module. TIME3D-IGGCAS will obtain the self-consistent calculated neutral composition, neutral density, neutral winds, and mid- and low-latitude ionospheric electric fields from GCITEM-IGGCAS model and provide the self-consistent calculated ion composition, ion and electron densities, ion and electron temperatures, and ion velocity vectors in the mid- and low-latitude ionospheres to GCITEM-IGGCAS model.
We use the two-way coupled GCITEM-TIME3D model in this work. The following simulations are performed at September equinox for low (F107, F107A = 70) or high (F107, F107A = 210) solar activity levels and geomagnetic quiet input with a cross cap potential of 20 kV and auroral particle precipitation with a hemispheric power of 10 GW. An IGRF geomagnetic field is used in these simulations. The initial conditions are from the MSIS00 and IRI2000 empirical models. The zonal mean states of neutral temperature/compositions from MSIS00 empirical model and the neutral temperature and density tides from TIMED/SABER observations are used at the low boundary (90-km altitude). These tides include the migrating diurnal tide, migrating semidiurnal tide, and non-migrating DE3 tide. The details of calculations of these tides from TIMED/SABER observations can be seen in Ren et al. (2011b) and Wan et al. (2012). However, to analyze the influence of DE3 tide on the equinoctial asymmetry, we will use DE3 tide at March equinox instead of that at September equinox in some simulations. To keep removing all the effect of the initial conditions, 15-model-day runs were made to obtain the presented results in all simulations.
Result and discussions
Figure 2c shows the latitudinal and altitudinal variations of δNe, and the solid lines in this figure represent the zero lines. Because most of the inputs and boundary conditions of these two simulations are the same as each other, δNe must be driven by the equinoctial asymmetry in DE3 tide and expresses the influence of DE3 tide on the equinoctial asymmetry of the zonal mean ionospheric electron density. As shown in Figure 2c, the value of δNe is mainly between -2% and 3%, and this asymmetry shows obvious latitudinal and altitudinal variations. We first focus on the altitudinal variation of δNe. At lower altitude (below about 170 km), δNe is mainly less than zero, and the March DE3 tide decreases the zonal mean electron density. At higher altitude, δNe is mainly larger than zero, and the March DE3 tide increases the zonal mean electron density. Figure 2d shows the altitudinal profiles of the latitudinal mean δNe (solid line), δNe at 0° (dashed line), δNe at 15° (EIA region, star line), and δNe at 45° (middle latitude, cycle line). Although four profiles show similar altitudinal variations in the MLT region, the altitudinal variations show large difference at higher altitude. The latitudinal mean δNe profile shows the global mean ionospheric response to DE3 tide, and its altitudinal variation is more consistent with the profile at middle latitude. The minimum of latitudinal mean δNe appears at about 93 km. Then, δNe increases with altitude and shows a peak near 130 km. Above 130 km, δNe begins to decrease with altitude again and shows the second minimum near 160 km. Above 160 km, δNe begins to increase with altitude again and shows a peak near 265 km. Above 265 km, δNe keeps decreasing with altitude.
Previous research implied that the ionospheric equinoctial asymmetry may vary with solar activity level, and Wan et al. (2008, 2012) also suggested that the ionospheric WN4 varies with solar activity level. Hence, we also compare δNe at high solar activity level with that at low solar activity level. Figure 3c,d respectively shows the latitudinal and altitudinal variations of the zonal mean ionospheric electron densities for low solar activity level and for high solar activity level, and the dashed lines in Figure 3a,b respectively shows the longitudinal averaged diurnal variation of the vertical drift at dip equator at the height of 320 km for low solar flux level and the vertical drift asymmetry driven by DE3 tide. The solid lines in Figure 3c,d represent the zero lines. At lower ionosphere, δNe at high solar activity level is mainly larger than that at low solar activity level, and the equinoctial asymmetry driven by DE3 tide decreases with the increase of solar activity. At higher mid- and high-latitude ionospheres, although δNe at high solar activity level also is mainly larger than that at low solar activity level, the equinoctial asymmetry driven by DE3 tide mainly increases with the increase of solar activity. At low-latitude ionosphere, the EIA structure also appears in the δNe at high solar activity level. Because the equator daytime upward drift for high solar flux level is stronger than that for low solar flux level, the spatial scale of EIA is larger than that at low solar activity level. However, because the daytime drift asymmetry for high solar flux level is weaker than that for low solar flux level, EIA of δNe at high solar activity level is obviously weaker than that at low solar activity level, and the EIA structure in equinoctial asymmetry mainly decreases with the increase of solar activity.
Summary and conclusion
In this paper, we simulate the influence of DE3 tide on the equinoctial asymmetry of the zonal mean states of ionospheric electron density with the GCITEM-IGGCAS model. Through respectively adding September DE3 tide and March DE3 tide at the low boundary, we simulate the distribution of ionospheric electron density and calculate the relative difference in zonal mean electron density. The influence of DE3 tide on the equinoctial asymmetry of the zonal mean electron density varies with latitude, altitude, and solar activity level. Compared with the density driven by the September DE3 tide, the March DE3 tide mainly decreases the lower ionospheric zonal mean electron density and mainly increases the electron density at higher ionosphere. In the low-latitude ionosphere, DE3 tide drives an EIA structure at higher ionosphere in the relative difference of zonal mean electron density, which may imply that DE3 tide affects the zonal mean equatorial vertical E × B plasma drifts. Although the lower ionospheric equinoctial asymmetry driven by DE3 tide mainly decreases with the increase of solar activity, the asymmetry at higher ionosphere mainly increases with solar activity. However, EIA in equinoctial asymmetry mainly decreases with the increase of solar activity.
This work is supported by the Chinese Academy of Sciences (KZZD-EW-01-2), National Important Basic Research Project (2011CB811405), National Science Foundation of China (41474133, 41322030, 41321003, 41131066), and Youth Innovation Promotion Association, CAS.
- Adebesin B, Adeniyi J, AAdimula I, Reinisch B: Equatorial vertical plasma drift velocities and electron densities inferred from ground-based ionosonde measurements during low solar activity. J Atmos Sol Terr Phys 2013, 97: 58–64.View ArticleGoogle Scholar
- Aruliah AL, Farmer AD, Fuller-Rowell TJ, Wild MN, Hapgood M, Rees D: An equinoctial asymmetry in the high-latitude thermosphere and ionosphere. J Geophys Res 1996, 101(A7):15713–15722. 10.1029/95JA01102View ArticleGoogle Scholar
- Bailey GJ, Su YZ, Oyama K-I: Yearly variations in the low-latitude topside ionosphere. Ann Geophys 2000, 18: 789–798. 10.1007/s00585-000-0789-0View ArticleGoogle Scholar
- Balan N, Otsuka Y, Fukao S, Bailey GJ: Equinoctial asymmetries in the ionosphere and thermosphere observed by the MU radar, J . Geophys Res 1998, 103(A5):9481–9486. 10.1029/97JA03137View ArticleGoogle Scholar
- Balan N, Kawamura S, Nakamura T, Yamamoto M, Fukao S, Oliver WL, Hagan ME, Aylward AD, Alleyne H: Simultaneous mesosphere–lower thermosphere and thermospheric F region observations using middle and upper atmosphere radar. J Geophys Res 2006, 111: A10S17. doi:10.1029/2005JA011487View ArticleGoogle Scholar
- Brahmanandam PS, Uma G, Liu JY, Chu YH, Latha Devi NSMP, Kakinami Y: Global S4 index variations observed using FORMOSAT-3/COSMIC GPS RO technique during a solar minimum year. J Geophys Res 2012, 117: A09322.Google Scholar
- Chen Y, Liu L, Wan W, Ren Z: Equinoctial asymmetry in solar activity variations of NmF2 and TEC. Ann Geophys 2012, 30: 613–622. 10.5194/angeo-30-613-2012View ArticleGoogle Scholar
- Forbes JM, Roble RG, Fesen C: Acceleration, heating, and compositional mixing of the thermosphere due to upward propagating tides. J Geophys Res 1993, 98(A1):311–321. 10.1029/92JA00442View ArticleGoogle Scholar
- Forbes JM, Zhang X, Palo S, Russell J, Mertens CJ, Mlynczak M: Tidal variability in the ionospheric dynamo region. J Geophys Res 2008, 113: A02310. doi:10.1029/2007JA012737Google Scholar
- Jin H, Miyoshi Y, Fujiwara H, Shinagawa H: Electrodynamics of the formation of ionospheric wave number4 longitudinal structure. J Geophys Res 2008, 113: A09307. doi:10.1029/2008JA013301Google Scholar
- Jones M Jr, Forbes JM, Hagan ME, Maute A: Impacts of vertically propagating tides on the mean state of the ionosphere-thermosphere system. J Geophys Res 2014, 119: 2197–2213. doi:10.1002/2013JA019744View ArticleGoogle Scholar
- Kawamura S, Balan N, Otsuka Y, Fukao S: Annual and semiannual variations of the midlatitude ionosphere under low solar activity. J Geophys Res 2002, 107(A8):1166. doi:10.1029/2001JA000267View ArticleGoogle Scholar
- Kil H, Oh S-J, Kelley MC, Paxton LJ, England SL, Talaat E, Min K-W, Su S-Y: Longitudinal structure of the vertical E × B drift and ion density seen from ROCSAT-1. Geophys Res Lett 2007, 34: L14110. doi:10.1029/2007GL030018View ArticleGoogle Scholar
- Lindzen RS: Thermally driven diurnal tide in the atmosphere. Q J Roy Meteorol Soc 1967, 93(395):18–42. 10.1002/qj.49709339503View ArticleGoogle Scholar
- Liu L, He M, Yue X, Ning B, Wan W: Ionosphere around equinoxes during low solar activity. J Geophys Res 2010, 115: A09307. doi:10.1029/2010JA015318Google Scholar
- Mukhtarov P, Pancheva D, Andonov B: Global structure and seasonal and interannual variability of the migrating diurnal tide seen in the SABER/TIMED temperatures between 20 and 120 km. J Geophys Res 2009, 114: A02309. doi:10.1029/2008JA013759Google Scholar
- Oberheide J, Forbes JM: Tidal propagation of deep tropical cloud signatures into the thermosphere. Geophys Res Lett 2008, 35: L04816. doi:10.1029/2007GL032397Google Scholar
- Oberheide J, Forbes JM, Hausler K, Wu Q, Bruinsma SL: Tropospheric tides from 80 to 400 km: propagation, interannual variability, and solar cycle effects. J Geophys Res 2009, 114: D00I05. doi:10.1029/2009JD012388Google Scholar
- Pancheva D, Mukhtarov P: Strong evidence for the tidal control on the longitudinal structure of the ionospheric F-region. Geophys Res Lett 2010, 37: L14105. doi:10.1029/2010GL044039View ArticleGoogle Scholar
- Ren Z, Wan W, Wei Y, Liu L, Yu T: A theoretical model for mid- and low-latitude ionospheric electric fields in realistic geomagnetic fields. Chin Sci Bull 2008, 53(24):3883–3890. 10.1007/s11434-008-0404-4Google Scholar
- Ren Z, Wan W, Liu L: GCITEM-IGGCAS: A new global coupled ionosphere-thermosphere-electrodynamics model. J Atmos Sol Terr Phys 2009, 71(17&18):2064–2076.View ArticleGoogle Scholar
- Ren Z, Wan W, Liu L, Xiong J: Intra-annual variation of wave number 4 structure of vertical E × B drifts in the equatorial ionosphere seen from ROCSAT-1. J Geophys Res 2009, 114: A05308. doi:10.1029/2009JA014060Google Scholar
- Ren Z, Wan W, Xiong J, Liu L: Simulated wave number 4 structure in equatorial F-region vertical plasma drifts. J Geophys Res 2010, 115: A05301. doi:10.1029/2009JA014746Google Scholar
- Ren Z, Wan W, Liu L, Chen Y, Le H: Equinoctial asymmetry of ionospheric vertical plasma drifts and its effect on F-region plasma density. J Geophys Res 2011, 116: A02308. doi:10.1029/2010JA016081Google Scholar
- Ren Z, Wan W, Liu L, Xiong J: Simulated longitudinal variations in the lower thermospheric nitric oxide induced by nonmigrating tides. J Geophys Res 2011, 116: A04301. doi:10.1029/2010JA016131Google Scholar
- Ren Z, Wan W, Liu L, Le H: TIME3D-IGGCAS: A new three-dimension mid- and low-latitude theoretical ionospheric model in realistic geomagnetic fields. J Atmos Sol Terr Phys 2012, 80: 258–266.View ArticleGoogle Scholar
- Ren Z, Wan W, Liu L, Xiong J: Simulated longitudinal variations in the E-region plasma density induced by non-migrating tides. J Atmos Sol Terr Phys 2012, 90: 68–76.View ArticleGoogle Scholar
- Ren Z, Wan W, Xiong J, Liu L: Simulated equinoctial asymmetry of the ionospheric vertical plasma drifts. J Geophys Res 2012, 117: A01301. doi:10.1029/2011JA016952Google Scholar
- Richmond AD: Ionospheric electrodynamics using magnetic apex coordinates. J Geomagn Geoelectr 1995, 47(1):191–212.View ArticleGoogle Scholar
- Sripathi S, Kakad B, Bhattacharyya A: Study of equinoctial asymmetry in the equatorial spread F (ESF) irregularities over Indian region using multi-instrument observations in the descending phase of solar cycle 23. J Geophys Res 2011, 116: A11302. doi:10.1029/2011JA016625View ArticleGoogle Scholar
- Titheridge JE: The electron content of the southern mid–latitude ionosphere, 1965–1971. J Atmos Terr Phys 1973, 35: 981–1001. 10.1016/0021-9169(73)90077-9View ArticleGoogle Scholar
- Titheridge JE, Buonsanto MJ: Annual variations in the electron content and height of the F layer in the Northern and Southern hemispheres, related to neutral composition. J Atmos Terr Phys 1983, 45: 683–696. 10.1016/S0021-9169(83)80027-0View ArticleGoogle Scholar
- Unnikrishnan K, Nair RB, Venugopal C: Harmonic analysis and an empirical model for TEC over Palehua. J Atmos Solar-Terr Phys 2002, 64: 1833–1840. 10.1016/S1364-6826(02)00187-6View ArticleGoogle Scholar
- Wan W, Liu L, Pi X, Zhang M-L, Ning B, Xiong J, Ding F: Wavenumber-four patterns of the total electron content over the low latitude ionosphere. Geophys Res Lett 2008, 35: L12104. doi:10.1029/2008GL033755View ArticleGoogle Scholar
- Wan W, Xiong J, Ren Z, Liu L, Zhang M-L, Ding F, Ning B, Zhao B, Yue X: Correlation between the ionospheric WN4 signature and the upper atmospheric DE3 tide. J Geophys Res 2010, 115: A11303. doi:10.1029/2010JA015527View ArticleGoogle Scholar
- Wan W, Ren Z, Ding F, Xiong J, Liu L, Ning B, Zhao B, Li G, Zhang M-L: A simulation study for the couplings between DE3 tide and longitudinal WN4 structure in the thermosphere and ionosphere. J Atmos Sol Terr Phys 2012, 90: 52–60.View ArticleGoogle Scholar
- Wu Q, Ortland DA, Foster B, Roble RG: Simulation of nonmigrating tide influences on the thermosphere and ionosphere with a TIMED data driven TIEGCM. J Atmos Sol Terr Phys 2012, 90: 61–67.View ArticleGoogle Scholar
- Zhang S-R, Holt JM, Zalucha AM, Amory–Mazaudier C: Midlatitude ionospheric plasma temperature climatology and empirical model based on Saint Santin incoherent scatter radar data from 1966 to 1987. J Geophys Res 2004, 109: A11311. doi:10.1029/2004JA010709View ArticleGoogle Scholar
- Zhao B, Wan W, Liu L, Mao T, Ren Z, Wang M, Christensen AB: Features of annual and semiannual variations derived from the global ionospheric maps of total electron content. Ann Geophys 2007, 25: 2513–2527.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.