1. Background: Galactic Cosmic Rays and Human Health
 In balloon flights in 1912–1913, Hess [1912, 1913] first measured a form of radiation that intensified with altitude. Millikan and Bowen  realized that this radiation couldn't come from the Sun since its intensity didn't vary from day to night, and thus called it “Cosmic Ray” radiation. The majority of this radiation is from galactic comic rays (GCRs), which are accelerated outside our Solar System, most likely by the strong shocks that result from supernova explosions in our galaxy. GCRs have enormous energies (100 MeV to 10 GeV) and near-light speeds. They continually bombard Earth's atmosphere, producing secondary particles and radiation through cascading high-energy collisions.
 Because most GCR radiation that penetrates into the near-Earth space environment is shielded by Earth's magnetic field, it does not pose a common health hazard. However, during space travel, GCRs are difficult to shield [Wilson et al., 1991; Cucinotta et al., 2001a, 2001b] since they produce secondary radiation in shielding and other material that is even more hazardous than the GCRs themselves. On long duration missions, GCR radiation is the primary health hazard to astronauts [Space Science Board, 1973; Space Studies Board, 1996; Cucinotta et al., 2001a].
 In the outer heliosphere, beyond ∼100 AU, the slowing of solar wind is thought to form a large magnetic barrier that shields out > 90% of the GCR radiation present in interstellar space [Florinski et al., 2003] at energies below ∼100 MeV. Because this reduction is so large, even a very small change in the shielding efficiency can have a large impact on the radiation environment in the solar system. However, because these regions have never been directly sampled or observed, there is great uncertainty about the physics of outer heliospheric shielding and its sensitivity to changes in the solar wind output and the local interstellar medium. A small fraction of GCRs penetrate into the heliosphere and propagate toward the Sun and planets. These residual GCRs are modulated by the solar wind's magnetic field in the inner heliosphere.
 Forbush  discovered the ∼11 year modulation of GCRs, which is explained by the transport of GCRs through heliospheric magnetic fields [Parker, 1965; Gleeson and Axford, 1968; Urch and Gleeson, 1972]. During periods of high solar activity (solar maximum), which occur each ∼11 years, the heliospheric magnetic field magnitudes increase due to a much larger rate of coronal mass ejections (CMEs) from the Sun. As a result, the solar wind's intense solar maximum magnetic field is more effective at sweeping GCRs out of the inner heliosphere, which causes a strong reduction of the GCR flux. A 22 year cycle is also evident: GCR flux maxima are typically sharply peaked for even numbered solar minima (when the large scale solar magnetic field is inward in the north; A < 0), whereas odd numbered solar minima (when the solar magnetic field is outward in the north; A > 0) show more broadly peaked GCR flux maxima. Over the 22 year solar magnetic cycle, known as the Hale cycle, the large-scale solar magnetic field reverses its polarity over the first solar activity cycle (∼11 years) and then returns to its original polarity in the next solar activity cycle. Most of the actual magnetic reversal happens over a 2–4 year period centered on solar maximum and is thought to be forced by the release of CMEs [Owens et al., 2007; Schwadron et al., 2008]. The solar magnetic reversal has a strong effect on the entry of GCRs into the heliosphere. Positively charged cosmic rays drift in from the heliospheric polar regions when the Sun's north polar field is directed outward (A > 0). In contrast, when the Sun's north polar field is directed inward (A < 0), positively charged cosmic rays drift inward along the heliospheric current sheet where they are scattered by irregularities in the current sheet, by coronal mass ejections and other large-scale interaction regions. Conversely, negatively charged cosmic rays (electrons) drift inward from directions opposite to the positively charged cosmic rays. Potgieter and le Roux  and Ferreira and Potgieter  approximately reproduce the ∼22 year cycle in the GCR fluxes using a time-dependent modulation model, which includes gradient, curvature, and current sheet drift effects [Jokipii et al., 1977] and diffusion parallel and perpendicular to the magnetic field.
 Large changes in the Local Interstellar Medium (LISM) have dramatic effects on the heliosphere and the radiation environment of the solar system. For example, a typical enhancement in the density of the local interstellar medium by a factor of 10 causes the entire heliosphere to shrink to about a quarter of its current size [Zank and Frisch, 1999], and increases the fluxes of GCRs at Earth by a factor of 2–6 [Scherer et al., 2002]. Such large changes in the LISM have certainly occurred in the past and will occur again in the future [Zank and Frisch, 1999].
 We do not currently have the observational knowledge required to understand how the local interstellar medium interacts with the heliosphere; continued observations such as those provided by the Interstellar Boundary Explorer mission (IBEX) [McComas et al., 2009] and the Voyager satellites [e.g., Stone et al., 2005; Richardson et al., 2008] are important for better understanding the radiation environment that must be traversed by astronauts for long missions to distant destinations such as Mars.
 On Earth, radioisotope 10Be provides a recent record of cosmic ray fluxes. The isotope is produced in Earth's upper atmosphere by spallation reactions of cosmic rays (CR) protons with energies higher than about 100 MeV and secondary neutrons with atmospheric nitrogen and oxygen [Masarik and Beer, 1999]. 10Be records in Antarctic ice show two prominent peaks 35,000 and 60,000 years ago, when the radioisotope production rate was about twice the current value for about 1500 and 2000 years, respectively [Raisbeck et al., 1987]. The 35,000 year peak is also present in marine sediment records [Cini Castagnoli et al., 1995; McHargue et al., 1995]. The fact that the 10Be radioisotope changes significantly with time shows that significant changes in cosmic ray fluxes have occurred in the past. But the cause for these changes remains controversial. The most accepted explanation is that the peaks are caused by geomagnetic magnetic field excursions, which is supported by the observed correlation between geomagnetic field strength and 10Be levels in marine sediment records [McHargue et al., 2000; Wagner et al., 2000]. However, this explanation would suggest that the 10Be enhancements should be uniform over the globe, which contradicts the large variations observed at high latitudes where geomagnetic effects are small. Further, Cini Castagnoli et al.  found that the 10Be level starts to increase some 2000 years before the drop in Earth's magnetic field, as reflected by the magnetization measured in the same core sample. Thus, many of the long-term changes in the cosmic ray fluxes incident on Earth may be due to effects external to the Earth; either due to changes in the shielding of the inner heliosheath or due to changes in the incident fluxes of GCRs from outside the heliosphere.
 Could the GCR fluxes in the heliosphere change rapidly in the future due to rapidly changing conditions in the LISM? For the period from 1700 AD to present-day, the 10Be record from ice cores can be compared with sunspot data [Beer et al., 1990]. The result is a remarkably steady anticorrelation, which demonstrates how changes in the solar wind's magnetic field over the solar cycle affect GCRs in the inner solar system. As described above, higher solar activity (when sunspot numbers are large) is correlated with increased interplanetary magnetic field strength, which, in turn, reduces the number of GCRs entering the inner heliosphere. A further implication of the strong and remarkably steady anticorrelation between sunspot number and GCR flux is that, over the last 300 years, the incident flux of GCRs on the heliosphere, and the shielding of GCRs in the inner heliosheath have been relatively constant.
 The primary question taken up in this paper is the dose rates induced by GCRs. The GCR environmental model of Badhwar and O'Neill [1991, 1992, 1993, 1994, 1996] is used to model the species-dependent GCR flux over time. The Badhwar-O'Neill model is based on the fitting of measured differential energy spectra to stationary solutions of the Fokker-Planck equation parameterized by an assumed form for the diffusion coefficient that varies as a function of the interplanetary modulation potential (measured in MV). The High-charge (Z) and Energy (HZE) Transport code (HZETRN) is solved using a one-dimensional, analytical formulation of the Boltzmann transport equation to describe the transport of GCRs through shielding and tissue. The effect of the long-range Coulomb force and electron interaction is treated as a continuous slowing-down process. Atomic (electronic) stopping power coefficients with energies above a few MeV/nucleon are calculated using Bethe's theory including Bragg's rule, Ziegler's shell corrections, and effective charge. Nuclear reaction cross sections are obtained from curve fits to quantum scattering calculations and total cross sections are obtained with a Ramsauer formalism. Nuclear fragmentation cross sections are calculated with a semiempirical abrasion-ablation fragmentation model, NUCFRG2, also developed at NASA Langley. The code transports heavy charged particles (Z = 1 through 26) in the incident GCR spectrum, and all of their reaction products, including secondary neutrons, through as many as 5 layers of shield materials and water, which is used as a surrogate for human tissue. The output from the code includes particle fluxes or fluences for all heavy charged particles and neutrons, absorbed dose, dose equivalents, and linear energy transfer (LET) spectra in water for any desired location in any of the shield materials.
 A 3-layer version of HZETRN 2005 has been configured for the Earth-Moon-Mars Radiation Environment Module (EMMREM) to calculate GCR dose and dose equivalent (L. Townsend et al., Transmission of solar energetic particles and galactic cosmic rays through the Mars atmosphere, submitted to Space Weather, 2009). The code has been used to develop a look-up table of daily effective dose, organ doses and dose equivalents behind varying thicknesses of aluminum shielding. GCR flux input into the calculations is taken from the Badhwar-O'Neill GCR model using modulation potentials ranging from the most highly probable solar minimum (400 MV) to solar maximum conditions (1800 MV) in the solar cycle. This model is the standard one used for space operations at the Space Radiation Analysis Group (SRAG) at NASA Johnson Space Center. A look-up table is used because the large spread in interplanetary magnetic field conditions, large numbers of GCR ion species and their many reaction product secondary particles must be transported through as much as 500 g/cm2 of shield materials. Calculations of such complex spectra at such depths take approximately half a day for each possible spectrum and cannot be carried out in near real time simulations. Since GCR intensities change very little from day to day, daily dose and dose equivalent estimates are sufficient.