SEPARATION OF THE RIBBON FROM GLOBALLY DISTRIBUTED ENERGETIC NEUTRAL ATOM FLUX USING THE FIRST FIVE YEARS OF IBEX OBSERVATIONS

The Interstellar Boundary Explorer (IBEX) mission has two sensors, IBEX-Lo (Fuselier et al. 2009b) and IBEX-Hi (Funsten et al. 2009a) that measure energetic neutral atoms (ENAs) with energies of ∼10 eV to 2 keV and ∼300 eV to 6 keV, respectively (McComas et al. 2009a, 2009b). The global ENA maps from IBEX characterize the interactions of the solar wind with the local interstellar medium. The first global maps of the heliosphere in ENAs (McComas et al. 2009a; Schwadron et al. 2009a; Fuselier et al. 2009a; Funsten et al. 2009b) showed the presence of a narrow ribbon (∼20°–40° wide) of elevated emissions that forms a circular arc roughly centered on ecliptic coordinates (long., lat.) ∼(2192 ± 13, 399 ± 23; Funsten et al. 2013). Comparisons between models of the outer heliosheath and the ribbon suggested that LISM magnetic field in the outer heliosheath is roughly perpendicular to the IBEX ribbon (or where B · r = 0, where r is the radial line-of-sight (LOS) and B is the interstellar magnetic field, McComas et al. 2009a; Schwadron et al. 2009a; Ratkiewicz et al. 2012). The center of the ribbon can potentially be considered as a proxy for the direction of the local interstellar magnetic field direction (Grygorczuk et al. 2011). The ribbon center direction is within ∼33°  ±  20° of the magnetic field direction derived from interstellar polarization data using stars within 40 pc (Frisch et al. 2012; Frisch & Schwadron 2013). Recently, (Schwadron et al. 2014) showed that the anisotropy maps of high-energy (TeV) cosmic rays provide independent confirmation of the interstellar magnetic field orientation inferred from the IBEX ribbon center. Therefore, mounting evidence shows that the center of the ribbon is the direction of interstellar magnetic field in the outer heliosheath.

The ribbon flux is notably superimposed on a slowly varying ENA flux that is referred to as the globally distributed flux (GDF) and is likely a separate emission population. The purpose of this paper is to utilize a method previously developed (Schwadron et al. 2011) to separate the GDF from the ribbon and discuss the properties of the two ENA populations. The first separation of the IBEX ribbon from the GDF was performed using the first year of IBEX-Hi data for the energy range from 0.7 to 4.3 keV. This study builds on (Schwadron et al. 2011) by performing ribbon separation using the first five years of IBEX-Hi and IBEX-Lo data covering a broadened energy range from 0.2 to 4.3 keV.

The GDF exhibits broad angular distributions far more consistent with original predictions of ENA maps from models (Prested et al. 2008; Schwadron et al. 2009a, 2009b). This led to the suggestion that the GDF most likely forms from charge exchange between interstellar neutrals and plasma in the inner heliosheath (McComas et al. 2009a; Schwadron et al. 2009a, 2011; Fuselier et al. 2009a; Funsten et al. 2009b). Therefore, maps of GDF are used to study properties of the inner heliosheath (Livadiotis et al. 2012). Maps of the IBEX ribbon are often used to test models of ribbon formation (see McComas et al. 2010, 2014b; Schwadron et al. 2011), to test the properties of the interstellar cloud surrounding the heliosphere (Frisch et al. 2010; Heerikhuisen et al. 2010, 2014; Ratkiewicz et al. 2012; Schwadron et al. 2014), and in comparison with observations of the interstellar magnetic field (e.g., Frisch et al. 2012; Frisch & Schwadron 2013). The maps of the GDF and the ribbon provided here will be released as new data products to the heliophysics community, allowing comparison between results of IBEX and models for both the ribbon and the inner heliosheath.

The starting point for our analysis is a set of global maps that are a statistically weighted combination of the first five years ENA sky maps (McComas et al. 2014a), shown in Figure 1. These maps are Compton–Getting corrected to the central energy of the sensor at every energy step in the inertial reference frame at 1 AU. Survival probability corrections are also applied. Therefore, the maps represent incident ENA fluxes from ∼100 AU (i.e., the outer heliosphere). The ENA maps are formed from data taken only when the sensors view the heliosphere in the ram direction (with a positive prograde component). For this "ram viewing," the Compton–Getting correction acts to suppress noise in the ENA fluxes (McComas et al. 2012). All maps have also been filtered according to signal-to-noise (S/N, ratio of flux to the standard deviation) such that any pixel with S/N <3 is zeroed.

Zoom In Zoom Out Reset image size

The maps in Figure 1 include the Compton–Getting and survival probability corrected IBEX-Lo ENA fluxes at 0.2 and 0.4 keV, also including only ram viewing. The IBEX-Lo maps are incomplete covering only the region from −30° to −180° in ecliptic longitude as the result of including only ram viewing and excluding time periods of high background when IBEX moves through the terrestrial magnetosphere.

IBEX-Lo maps have been corrected by subtraction of the sputtering contributions from O and H, as detailed in the Appendix of (McComas et al. 2014a). The sputtering factors have been calculated using a combination of calibration and flight data. We also include corrections to IBEX-Lo geometric factors due to changes in the IBEX-Lo post acceleration (PAC) voltage and changes in data throughput from the instrument to the central electronics unit (throughput is defined as the percentage of measured events that survive in the data stream to the central electronics unit). The average throughput was 80% in IBEX-Lo ESA 5, 94% in IBEX-Lo ESA 6, and 96% in ESA 7 and 8 prior to orbit 168. After orbit 168, the throughput was improved to ∼100% by removing unused housekeeping in the data stream. In addition, the PAC voltage was lowered in orbit 177 to exclude backgrounds, which required subsequent adjustment to the IBEX-Lo geometric factors. Finally, the IBEX-Lo ESA 5 subtracts a small background rate of 0.005 ± 0.001 s⁻¹ (A. Galli et al. 2014, in preparation; there is no background correction applied for IBEX-Lo ESA 6). Background rates are also subtracted from IBEX-Hi data, as detailed by (McComas et al. 2014a).

We have applied the same techniques as detailed in (Schwadron et al. 2011) to separate the ribbon from the GDF. The technique proceeds by first rotating the global maps of IBEX into a frame with a polar axis near the ribbon center, taking advantage of the ribbon circularity (eccentricity ∼0.3, Funsten et al. 2013). This rotation makes it relatively straightforward to perform fits to the ribbon and then use these fits to separate out the ribbon from the GDF ENA populations. The separation technique is entirely independent at different energy steps. The method of ribbon separation is not strongly dependent on the choice of ribbon center. For convenience, we have adopted a ribbon center at ecliptic longitude 221° and ecliptic latitude 39°, which is within 2° from the ribbon center derived by (Funsten et al. 2013). Figures 2–4 show the separated ribbon flux maps and Figures 5–7 shows the GDF maps in ecliptic, equatorial, and galactic coordinates.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

We fit the GDF as a function of energy to determine the spectral index, γ, where the differential flux in each pixel is approximated J∝E^γ and E is energy. The spectral index fitting uses a standard least squares method discussed by (Schwadron et al. 2011). Characterizing the differential flux using a spectral index in each pixel has been shown to be a reasonable approximation over most of the map. However, near the poles, we often observe broken ENA energy spectra (Dayeh et al. 2012). The spectral indices in the GDF are shown in Figure 8.

Zoom In Zoom Out Reset image size

We have calculated the pressure of plasma protons forming ENAs integrated over LOS where they are formed:

$$\begin{eqnarray} P_\mathrm{stationary} \cdot \mathrm{LOS} & = & \frac{2\pi m^2}{3 n_{\rm H}} \int _{E_\mathrm{min}}^{E_\mathrm{max}} \frac{dE}{E} \frac{j_\mathrm{ENA}(E)}{\sigma (E)} (|v|)^3. \end{eqnarray} \tag{ 1 }$$

The limits of integration extend over the range of energies for which we have direct observational information from ENAs. The ENA flux is j_ENA(E) at energy E with corresponding particle speed v. This "LOS-integrated pressure" is calculated in the inertial reference frame (Figure 9). Note that in Figure 8 and in the top panel of Figure 9, we include both IBEX-Lo and IBEX-Hi data. Because IBEX-Lo data includes only regions centered on the heliosphere nose (defined by the upwind direction of the interstellar He flow), we observe a break in intensity at ecliptic longitudes of −30° and 180°.

Zoom In Zoom Out Reset image size

Figure 8 reveals what has been pointed out in the first published maps (McComas et al. 2009a; Schwadron et al. 2009a; Fuselier et al. 2009a; Funsten et al. 2009b): the spectrum tends to harden at the poles and is softest near the tail. Originally, these trends were thought to be a property of both the ribbon and the GDF. However, subsequent work (Schwadron et al. 2011) where the GDF was studied separately showed that this property of the spectrum is driven by the GDF, even in the regions where the ribbon is observed. Understanding the detailed relationship between the energy spectrum and the varying global properties of the heliosheath remains an issue that has been studied only marginally using sophisticated modeling and theory. Recent work (e.g., Desai et al. 2014; Fuselier et al. 2014) has demonstrated both the ability of models to reproduce observations in the Voyager 1 direction and the direct extension of IBEX spectra into the much higher energies observed in situ by Voyager 1.

The GDF LOS-integrated pressure has a pronounced asymmetric enhancement near the nose (Figure 9). The asymmetric enhancement is shifted to the south and, near the ecliptic, the enhancement is shifted to the port direction (starboard and port are nautical terms detailed by McComas et al. 2013b).

Figure 10 shows the GDF centered on the interstellar downwind direction. We observe a clear signature of the tail along with the asymmetric lobes of ENA flux depletions (McComas et al. 2013b). Figure 11 shows the spectral indices of the GDF (top panel) and the LOS-integrated pressure (bottom panel) centered on the downwind direction. There is a LOS-integrated pressure enhancement shifted ∼10° to the starboard direction, yet the spectral index remains similar throughout the tail region including the side-lobes where the ENA flux is depleted. The softer spectrum near the tail region and in the lobes is likely due to the slower solar wind that, on average, emanates from the Sun in these regions (McComas et al. 2013b). The enhanced LOS-integrated pressure near the core of the tail is likely the result of the larger LOS that develops along the exhaust of heliosheath plasma. This suggests further that the large difference between the two tail lobes (i.e., the larger ENA flux depletion in port lobe) is driven by physical asymmetry of the lobes.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Based on the LOS-integrated pressure (Figure 9) and the spectral index map (Figure 11), we define the following structures ordered by ecliptic longitude:

• 1.  
  the nose region of pressure enhancement between ecliptic longitudes 160° and 320°,
• 2.  
  the center tail region between ecliptic longitudes 50° and 130°,
• 3.  
  the port lobe region between ecliptic longitudes −30° and 50°,
• 4.  
  the starboard lobe region between ecliptic longitudes 130° and 180°.

The definition of these regions becomes useful primarily in identifying relationships between the longitudinal variation of the ribbon and the GDF.

Figure 12 (top) plots the total differential flux (no ribbon separation has been applied) of ENAs as a function of ribbon latitude in the frame of reference with a polar axis at the ribbon center. The normalized differential flux (we have multiplied each latitudinal distribution by a constant to normalize the maximum to 1.0) of the separated ribbon in this frame (bottom panel) allows direct comparison of the ribbon width and ribbon center as a function of energy. We observe broadening of the ribbon at higher energies (2.7 and 4.3 keV) and also at lower energies (0.2 and 0.4 keV). The ribbon appears most narrow at energies of 0.7 and 1.1 keV, which are energies associated with characteristic slow solar wind speeds of 370 and 460 km s⁻¹, respectively.

Zoom In Zoom Out Reset image size

Figures 13 and 14 show the general characteristics of the ribbon. Shown in Figure 13 is the ecliptic latitude and ecliptic longitude of the ribbon as a function of the ribbon longitude in the rotated frame. This figure highlights the remarkable organization of the ribbon that is consistent across all energies observed. Shown in Figure 14 is the latitude of the ribbon peak in the rotated reference frame (top panels) and the FWHM (bottom panels) of the ribbon. The panel columns organized according to energy. General ordering of ribbon characteristics as a function of ecliptic longitude reveals the effects of heliosheath structure on the ribbon.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The ribbon latitude in the rotated reference frame reveals a number of organizing features. Figure 14(b), for example, at energies 0.7–1.7 keV shows that the ribbon appears almost at the same rotated latitude of ∼15° across the entire structure. If the ribbon were a great circle, it would appear at 0°. Instead the ribbon is shifted in latitude toward the ribbon center (the ribbon center in this rotated frame is at latitude +90°).

Consider the ribbon as a structure that lies along the surface where the radial direction (along the LOS) is perpendicular to the local interstellar magnetic field. In this case, the conical surface in Figure 15 represents the tangent surface containing the interstellar magnetic field at the location of the offset circle (the ribbon). Therefore the latitude of the ribbon may represent the deflection of the interstellar magnetic field due to draping at the location of the ribbon.

Zoom In Zoom Out Reset image size

In both Panels (a) and (b) of Figure 14 for energies from 0.2–1.7 keV, we find an increase of ribbon latitude near the nose, suggesting increased draping in this region. In Figures 14(d) and (e), we observe a characteristic broadening of the ribbon near the nose that appears coincident with the increase in ribbon latitude. One interpretation is that this broadening may be the result of the changing deflection due to different amounts of draping along the LOS.

In Panels (c) and (f) of Figure 14 at 2.7 and 4.3 keV energies, we find broadening near the nose, but the ribbon latitude increases in an asymmetric manner. In particular, the ribbon latitude appears to increase only in the port side of the nose. Since the ribbon center is on the starboard side of the nose, the asymmetry may reflect interaction between the ribbon and the interstellar flow.

At or near ∼120°, we observe a decrease in ribbon latitude and a decrease in the ribbon width. Note that this ecliptic longitude is near both the tail and the region where the ribbon crosses the equator in the ecliptic (the ecliptic latitude of the ribbon is near 0°). The decrease in ribbon latitude indicates that the structure changes into something closer to a great circle. This shift and the decrease in ribbon width are both indicators that the ribbon is formed in a region where the interstellar field is closer to a planar structure in the asymptotic limit of no draping. The reduction of the ribbon width may also signify the departure from draping and return to an ordered planar structure along the LOS. These are characteristics that may naturally exist far from the Sun in the tail region.

In this section, we discuss implications of the separated maps of the ribbon and GDF. The discussion is organized first in terms of implications of the GDF for the heliospheric tail and lobes (Section 3.1) and for the region near the nose (Section 3.2). The considerations in these two first subsections are used to develop a basic picture for the global ecliptic structure of the heliosphere (Section 3.3). We then discuss implications of the separated ribbon maps for the source of the ribbon (Section 3.4).

3.1. Heliospheric Tail and Lobes

(McComas et al. 2012, 2013b) demonstrated for the first time based on the first three years of IBEX data the existence of a heliospheric tail centered close (within 15°) to the downwind direction. The spectral index of the energy spectrum is γ ≈ −2.3. (Figure 11, top) and is roughly uniform throughout the tail region. However, the pressure integrated across LOS (Figure 11, bottom) seems to peak slightly south of the ecliptic (∼ − 10° ecliptic latitude) at an ecliptic longitude of ∼90°, about 10° from the downwind direction (79° ecliptic longitude). Centered at about 15° ecliptic longitude we observe a deep reduction in ENA flux in the port lobe and a shallower ENA flux reduction from 140° to 180° ecliptic longitude in the starboard lobe.

We consider the implications of the pressure integrated across LOS for global heliospheric structures near the tail and in the two lobes. Recall that the LOS-integrated pressure as shown in Figure 11 is in the heliospheric frame, not that of the plasma. The formula for converting the LOS-integrated pressure to the plasma frame was developed by (Schwadron et al. 2011) and found to depend on the downstream flow speed and the spectral index.

(Schwadron et al. 2011) developed and applied a simple mass-loading model that provides necessary context for detailing the plasma environment of the tail, and thereby, interpreting the integrated LOS pressures in the GDF. The model based on (Isenberg 1987) applies conditions near 1 AU and integrates the solar wind plasma properties out through the heliosphere as new pickup ions created from interstellar neutral atoms mass-load the solar wind. The effect of mass-loading reduces the solar wind speed with increasing distance from the Sun. Here, we apply the mass loading model assuming a solar wind particle flux at 1 AU of 3.5 × 10⁸ cm⁻² s⁻¹, a solar wind speed at 1 AU of 450 km s⁻¹, and a neutral hydrogen density of 0.07 cm⁻³. These values are derived from solar wind conditions consistent with a period of time roughly 8 years ago, in the time frame of 2006. The 8 year delay time is associated with the time required for solar wind to move back into the heliotail (1.5 years to 150 AU in supersonic solar wind, 4 years for plasma to move an additional 100 AU in the subsonic inner heliosheath, and 2.5 years for ENAs to travel back to IBEX).

The pressure in the frame of reference moving with the plasma at radial speed u_R is

$$\begin{eqnarray} &&P_\mathrm{plasma, R} = P_\mathrm{internal} + P_\mathrm{ram}, \end{eqnarray} \tag{ 2 }$$

where

$$\begin{eqnarray} P_\mathrm{internal} & = & \frac{4\pi m}{3} \int _{v_\mathrm{p min}}^{v_\mathrm{p max}} dv_p v_p^4 f_p(\mathbf {v}_p) \mathrm{, \;and} \end{eqnarray} \tag{ 3 }$$

$$\begin{eqnarray} P_\mathrm{ram} & = & 4\pi m u_R^2 \int _{v_\mathrm{p min}}^{v_\mathrm{p max}} dv_p v_p^2 f_p(\mathbf {v}_p) \end{eqnarray} \tag{ 4 }$$

are the internal plasma pressure and ram pressure, respectively. The limits of integration extend over the portion of the plasma distribution function (f_p) for which we have direct observational information from ENAs. The distribution function of ENAs is f_{p, ENA}(v_p) = f_p(v_p)/[n_Hσ(E_p)LOS], where n_H is the neutral H density, and σ(E_p) is the charge-exchange cross section at energy $E_p = mv_p^2/2$. Here, the subscript p indicates the plasma frame. The velocity of a particle in the plasma frame, v_p, is related to the velocity in the inertial (or observer) rest frame, v_o, by the following relation, $\mathbf {v}_p = \mathbf {v}_o - u_R\hat{e}_R$, where the unit radial vector is $\hat{e}_R$. The particle velocity is directed radially inward toward the observer so that $v_p^2 = (|v_o| + u_R)^2$. The observed differential flux, j_{o, ENA}(E_o) = f_{o, ENA}2E_o/m². Incorporating these transformations, we express the internal and ram pressure as follows:

$$\begin{eqnarray} P_\mathrm{internal} \cdot \mathrm{LOS} & = & \frac{2\pi m^2}{3 n_{\rm H}} \int _{E_\mathrm{min}}^{E_\mathrm{max}} \frac{dE_o}{E_o} \frac{j_{o,{\mathrm {ENA}}}(E_o)}{\sigma (E_p)} \frac{(|v_o| + u_R)^4}{|v_o|} \nonumber\\ \end{eqnarray} \tag{ 5 }$$

$$\begin{eqnarray} P_\mathrm{ram} \cdot \mathrm{LOS} & = & \frac{2\pi m^2 u_R^2}{ n_{\rm H}} \int _{E_\mathrm{min}}^{E_\mathrm{max}} \frac{dE_o}{E_o} \frac{j_{o,{\mathrm {ENA}}}(E_o)}{\sigma (E_p)} \frac{(|v_o| + u_R)^2}{|v_o|}. \nonumber\\ \end{eqnarray} \tag{ 6 }$$

We use the mass loading model and the observed properties of ENA flux to derive a correction factor for the LOS-integrated stationary pressure shown in Figure 11. When analyzing the tail it is important to recognize that the plasma keeps flowing down-tail. The LOS in this case (near the tail) is the distance over which the primary plasma from the solar wind becomes neutralized through charge-exchange. The charge-exchange process merely transforms the originally hot solar wind plasma into a cooler and slower plasma dominated by interstellar atoms. (Schwadron et al. 2011) used the term cooling length to refer to this characteristic distance over which charge-exchange cools and slows the plasma. Because the material in the tail is not diverted or stagnated, the LOS-integrated internal pressure provides the basis for comparison to the mass-loading model.

Equation (5) determines the correction factor that multiplies the observed LOS-integrated stationary pressure to determine the internal LOS-integrated pressure. The correction factor depends on the spectral index (γ = −2.3) in the observed ENA spectrum and the plasma flow speed downtail. The mass loading model results in a downtail flow speed of u_R ≈ 110 km s⁻¹ for a wide-range of possible termination shock (TS) radii from 100 to 170 AU. Applying these values, we find that the correction factor is C = 2.42. Based on the results in Figure 11, this implies the P_{internal, tail} · LOS ∼ 73 pdyne cm⁻² AU in the core tail, P_{internal, port} · LOS ∼ 36 pdyne cm⁻² AU in the port lobe and P_{internal, star} · LOS ∼ 48 pdyne cm⁻² AU in the starboard lobe. Note that near the tail we have the benefit of observations down to 0.7 keV, but not all the way down to 0.2 keV. As a result, these LOS-integrated pressures are lower limits for the actual plasma pressures.

The internal pressure of the inner heliosheath plasma (beyond the TS) is shown in red (Figure 16) as a function of TS radius from the mass loading model. Given the internal pressure, and the values for the LOS-integrated internal pressures, we then solve for the LOS length as a function of TS radius near the tail (dark blue), in the port lobe (purple) and in the starboard lobe (light blue).

Zoom In Zoom Out Reset image size

The cooling length, given by

$$\begin{eqnarray} &&l_c = \frac{u_R}{v_p \sigma n_{\rm H}}, \end{eqnarray} \tag{ 7 }$$

sets a maximum for the possible LOS-length. The cooling length is shown as a function of energy in Figure 17. Based on the spectral index, the pressure integrated over LOS is dominated by the lowest energies observed (0.7 keV near the tail). This suggests a limiting LOS of 133 AU near the tail. Based on this constraint and the results shown in Figure 16, we estimate a TS radius near the tail of ∼163 AU.

Zoom In Zoom Out Reset image size

The lower ENA emissions and LOS-integrated pressures in the port and starboard lobes suggest that the inner heliosheath thickness (d_Hsh) is reduced in these regions. Assuming that the internal pressure is roughly uniform across the tail and in the lobes, we find d_Hsh ∼ 66 AU in the port lobe and d_Hsh ∼ 89 AU in the starboard lobe.

In addition to LOS lengths, the internal pressure in the tail region provides an important measurement. We find P_{internal, tail} ∼ 0.55 pdyne cm⁻², which is slightly higher (by a factor of 1.5) than the magnetic pressure induced by a 3 μG magnetic field. A small 23% compression of the LISM magnetic field would induce a comparable internal pressure in the tail.

One of the consequences of the reduction of cooling length with increasing energy (Figure 17) is that the tail feature should fade with increasing energy. Figure 11 shows clearly that the tail feature fades for energies greater than 1.7 keV. At these energies, the cooling length decreases to less than 100 AU and approaches the 89 AU inner sheath thickness in the starboard lobe. The consistency of the length-scales validates the developing picture of the tail region.

3.2. Heliospheric Nose

We observe a LOS-integrated stationary pressure of the GDF (see Figure 9) near the nose given by ∼42 pdyne AU cm⁻². At the location of Voyager 1 and Voyager 2, this LOS-integrated pressure is ∼33 pdyne AU cm⁻². For analysis near the location of V1, we apply the mass loading model assuming a 1 AU solar wind particle flux of 3 × 10⁸ cm⁻² s⁻¹, a solar wind speed at 1 AU of 400 km s⁻¹, and a neutral density of 0.08 cm⁻³. These values are derived from solar wind conditions consistent with a period of time roughly 4 years ago (in the time frame of 2010). The shorter transit time delay near the nose is due to the smaller distance of travel of solar wind ions from the Sun into the inner heliosheath. The slight enhancement of the neutral density compared to that in the tail region is due to reduced charge-exchange loss of neutrals since the path history of neutrals intersects a thin portion of the heliosheath prior to arrival near the nose.

The downstream plasma speed from the mass loading model near the nose is ∼100 km s⁻¹ for a range of TS radii from 100 to 200 AU. Given a spectral index in the differential energy flux of γ = −1.7, we find that the correction factor is C = 2, which multiplies the observed stationary LOS-integrated pressure (Figure 9) to deduce the LOS-integrated internal pressure, as discussed previously. Based on the results in Figure 9, this implies that P_{internal, V1} · LOS ∼ 66 pdyne AU cm⁻².

The trends in solar wind pressure and particle flux appear to be relatively stable over the five-year period for which maps were analyzed (McComas et al. 2013a), which implies that the distance to the termination shock should also remain roughly constant. When compared to the 2005 time frame when Voyager 1 crossed the termination shock, and adding about a year for transit time, the 1 AU solar wind dynamic pressure appears to be about 15% larger than in the 2009-2014 period during which IBEX observations were made. Given the 1/r² decrease in solar wind pressure, the 1 AU observations suggest that the termination shock should have moved in to ∼87 AU during the 2009-2014 time period (for comparison, see Webber & Intriligator 2011). Assuming a TS radius of 87 AU, the mass loading model yields a downstream internal pressure of 1.7 pdyne cm⁻². Dividing this pressure into P_{internal, nose} · LOS, we find a LOS ∼ 40 AU, which would suggest a heliopause near 127 AU. This estimate is somewhat larger the latest results from Voyager 1 indicating that it has moved into a region that is close to interstellar plasma at ∼121 AU (Gurnett et al. 2013).

Note that our calculation assumes a stationary termination shock and heliopause. In order to consider potential implications of time-dependent changes in the sheath, we re-ran the separation of the ribbon from the GDF for the year 5 maps (in 2013). This time-period is roughly consistent with the time period of the Voyager 1 121 AU crossing when correcting for solar wind transit time. Figure 18 shows the resulting LOS integrated pressure. When compared to the LOS-integrated pressure over the full five years of data (Figure 9), we observe the significant (∼19%) reduction of LOS-integrated pressure near the nose. Since the solar wind properties appear to be comparably stable over this time frame (2009–2014), the reduction in LOS-integrated pressure appears to be associated with inward motion of the heliopause and therefore a reduction in the LOS. Specifically, since the LOS-integrated pressure is reduced by ∼19%, we find a LOS of ∼32 AU. Given a TS at 87 AU, this would suggest a heliopause crossing at 119 AU. This is only 2 AU smaller than the distance for the actual Voyager 1 crossing.

Zoom In Zoom Out Reset image size

One result is particularly interesting when Figure 18 is compared with Figure 9. While the LOS-integrated pressure near the nose drops, the tailward LOS-integrated pressure remains comparatively stable. Over the period from 2010 to 2014, we observe a fairly stable solar wind ram pressure, suggesting that the drop in LOS-integrated pressure near the nose is due to a more significant decrease in the LOS as compared to the plasma pressure across the nose region.

The ram pressure in solar wind was higher prior to 2010 than it was in the period from 2010 to 2014. This suggests that the inner heliosheath near the nose underwent inward motion to adjust to changing solar wind conditions. This inward motion must have occurred first at the termination shock, which responds immediately to local changes in the upstream supersonic solar wind. There is inherent latency in the inner heliosheath, which responds globally to average changes in solar wind over multi-year timescales since it is a subsonic medium. The reduction in solar wind pressure from 2006 to 2010 would have caused a steady inward motion of the termination shock (e.g., Richardson & Wang 2011), and a several-year latency in the inward motion in the heliopause. Due to the stability in solar wind ram pressure from 2010-2014, we expect that the termination shock would have largely stopped moving inward, while the heliopause continued to move inward over this time frame. The heliopause likely moved inward over Voyager 1 during the crossing.

3.3. Global Ecliptic Structure of the Longitudinally Asymmetric Heliosphere

Based on our analysis in the previous two subsections, we form a rough picture (Figure 19) of the heliosphere (termination shock and heliopause boundary) near the ecliptic plane. The termination shock distances and LOS thicknesses near the nose, tail and in the lobes are interpolated to illustrate the global structure. The direction of the interstellar magnetic field is given by the center of the IBEX ribbon and is projected into the ecliptic plane. The draping configuration is illustrative. This figure is approximate and its accuracy will evolve as the sophistication of GDF analysis improves. We emphasize that the LOS distance to the heliopause near the tail is essentially the charge-exchange cooling length. It is likely that the tail extends beyond this soft boundary, which sets a rough maximum for the LOS distance in this region.

Zoom In Zoom Out Reset image size

Results of modeling of the heliosheath near the nose (see Figure 7 in Pogorelov et al. 2011) also show pressure structure consistent with asymmetries in the GDF (e.g., Figure 9). Both the modeled magnetic and total pressure on the heliopause is at a maximum near the upwind direction, but to the south by ∼15° and to port by a similar angle. While the southward deflection of total pressure appears consistent with the GDF LOS-integrated pressure, the shift in the pressure to port is opposite to the starboard shift in pressure observed in the GDF. This difference may again be a reflection of the asymmetry in the LOS. There remains a need to utilize global models of the heliosphere in combination with maps of the GDF to develop a consistent global picture heliosheath thickness and TS radius.

3.4. The IBEX Ribbon

We have conducted the second analysis in which we have separated the IBEX ribbon from the GDF. With the benefit of 5 years of data from IBEX, we have resolved and analyzed full-sky maps for both the ribbon and GDF. We also now have the addition of ribbon separation involving IBEX-Lo data down to 0.2 keV. The more complete analysis reveals features that were not apparent in the first ribbon-separation analysis (Schwadron et al. 2011) yielding a wealth of new information about our global heliosphere:

• 1.  
  Global asymmetry. An important result from the analysis is the existence of an asymmetry in the global heliosphere observed both in the thickness of the heliosheath near the tail and the offset of the tail. The heliosheath appears to be thinner on its port side, and the tail is shifted toward the starboard direction. Both effects suggest a pressure enhancement on the port side, possibly a manifestation of compression and draping of the magnetic field. This result and our understanding of heliospheric asymmetry are ripe for further work involving global heliospheric modeling.
• 2.  
  Cooling-length limit for LOS. We have shown that the cooling-length limit for the LOS decreases with increasing energy. This is consistent with observations in Figure 5 showing that that tail signature erodes at energies greater than 1.7 keV where the cooling length becomes smaller than 100 AU.
• 3.  
  Broadening of ribbon in relation to magnetic draping. We observe a general trend in the ribbon (Figure 14) showing its width broadens near the nose of the heliosphere. This suggests the importance of magnetic draping that distorts the interstellar magnetic field, moving the ribbon to higher latitudes in the frame centered on the ribbon (Figure 14, bottom panel). Conversely, near the tail the ribbon appears to be most narrow and is shifted to the lowest latitudes in the ribbon-centered reference frame, consistent with a slight elliptical elongation of the ribbon found by Funsten et al. (2013).
• 4.  
  Time variability near the nose. We have also re-analyzed the properties near the nose of heliosphere based on the LOS-integrated pressure in ENAs observed there. Taking into account recent Voyager 1 observations, we develop a rough model of ENA emissions near the nose that demonstrates the importance of time variability in the heliosheath. In comparing the LOS-integrated pressures averaged over 5 years versus those in 2013, we find that the pressure in the inner heliosheath nose region has dropped significantly in 2013. Given the relative stability of the solar wind pressure from 2010 to 2014, our results suggest that the heliopause has been moving inward over this time frame, a latent response to the drop in solar wind pressure occurring from 2006 to 2010.
• 5.  
  Examination of ribbon models. The separated maps of the IBEX ribbon have allowed careful examination of the ribbon models. Generally, the variability in the properties including the central location of the ribbon as a function of energy suggest that the ribbon does not exist at a well defined physical location or at an interface. This constraint poses a significant problem for a number of the proposed models. The ribbon also exists as separable structure down to 0.2 keV. However, at these low energies, the ribbon both broadens and exhibits considerable variability in location as a function of ecliptic longitude.

The key products of our analysis are separated maps of the IBEX ribbon and the GDF. These maps will be the subject of the next IBEX data release (release 8). Generally, we observe the ribbon in every energy step analyzed from 0.2 keV up through 4.29 keV involving both IBEX-Lo and IBEX-Hi data. The resulting maps of the ribbon and the GDF have fundamental implications for the structure of our global heliosphere, and the structure, source and energy dependence of the IBEX ribbon.

We are deeply indebted to all of the outstanding people who have made the IBEX mission possible. This work was carried out as a part of the IBEX project, with support from NASA's Explorer Program. J.S., M.B., and M.A.K. were supported by the Polish National Science Centre (grant 2012-06-M-ST9-00455).