Cosmic-ray interaction data for designing biological experiments in space

Accepted Manuscript

Cosmic-ray interaction data for designing biological experiments in space T. Straume , T.C. Slaba , S. Bhattacharya , L.A. Braby PII: DOI: Reference:

S2214-5524(17)30013-5 10.1016/j.lssr.2017.04.002 LSSR 129

To appear in:

Life Sciences in Space Research

Received date: Revised date: Accepted date:

3 February 2017 17 April 2017 19 April 2017

Please cite this article as: T. Straume , T.C. Slaba , S. Bhattacharya , L.A. Braby , Cosmic-ray interaction data for designing biological experiments in space, Life Sciences in Space Research (2017), doi: 10.1016/j.lssr.2017.04.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Cosmic-ray interaction data for designing biological experiments in space T. Straume1*, T.C. Slaba2, S. Bhattacharya1, L.A. Braby3 1

NASA Ames Research Center, Moffett Field, CA 94035

2

NASA Langley Research Center, Hampton, VA 23681 Texas A&M University, College Station, TX 77843

CR IP T

3

*Corresponding author: Dr. Tore Straume, Space Biosciences Division, NASA Ames

AN US

Research Center, Mail Stop 236-7, Moffett Field, CA 94035. [email protected].

Abstract: There is growing interest in flying biological experiments beyond low-Earth orbit (LEO) to measure biological responses potentially relevant to those expected during a human mission to Mars. Such experiments could be payloads onboard precursor missions, including unmanned private-public partnerships, as well as small low-cost spacecraft (satellites)

M

designed specifically for "biosentinel" type missions. It is the purpose of this paper to provide physical cosmic-ray interaction data and related information useful to biologists who may be

ED

planning such experiments. It is not the objective here to actually design such experiments or provide radiobiological response functions, which would be specific for each experiment and biological endpoint. Nuclide-specific flux and dose rates were calculated using OLTARIS and

PT

these results were used to determine particle traversal rates and doses in hypothetical biological targets. Comparisons are provided between GCR in interplanetary space and inside

CE

the ISS. Calculated probabilistic estimates of dose from solar particle events are also presented. Although the focus here is on biological experiments, the information provided

AC

may be useful for designing other payloads as well if the space radiation environment is a factor to be considered.

Key Words: biosentinel, galactic cosmic rays, GCR, solar energetic particle events, SEP, SPE, nuclide-specific dose rates, nuclide-specific flux, particle track traversal rates, cell nucleus, track structure, International Space Station, ISS, interplanetary space, Mars, LEO

1

ACCEPTED MANUSCRIPT

Introduction

Understanding the impact of the space environment on biological systems is becoming particularly important now that extended human missions beyond low Earth orbit (LEO) are being planned (e.g., NASA 2016). The radiation environment in interplanetary space, including that expected during future human missions to Mars, is both quantitatively and

CR IP T

qualitatively different from that in LEO. The International Space Station (ISS) is a good test platform in LEO to evaluate biological responses to various space-flight factors, including microgravity. However, radiation-induced biological responses evaluated using experiments on the ISS have been difficult to quantify due to low dose rate, relatively few high-LET particles, and the challenges associated with obtaining suitable controls for comparison. This

AN US

is discussed in a later section "Interplanetary Space vs the International Space Station (ISS)".

Biological experiments as secondary payloads associated with unmanned precursor missions are under development to evaluate the space environment beyond LEO and return critical biological response information via telemetry (e.g., Bhattacharya et al. 2016). Such

M

"biosentinel" missions would be relatively low cost and provide flight opportunities, but are technically and scientifically challenging. It is not the objective here to design such

ED

experiments. That would be done by the experimenters planning them and would require radiobiological response information specific for the biological systems and endpoints to be studied. Rather, it is the purpose of this paper to provide physical cosmic-ray interaction data

PT

useful to biologists who may be planning such experiments, and who may not be familiar with

CE

these kinds of data.

The radiation environment in interplanetary space is complex and consists of galactic

AC

cosmic radiation (GCR), which is slowly varying during an 11-year solar cycle, and an occasional (and not yet predictable) solar energetic particle (SEP) event. Radiation dose rates from GCR based on model calculations vary by factors of 2 or more between solar minimum and solar maximum. The solar modulation data from ground-based measurements would suggest factors of 2.23 - 2.98 between 1955 and 2010 (Usoskin et al. 2011). The largest dose rate from GCR occurs during solar minimum because the Sun’s diminished solar wind permits more charged particles from our galaxy to enter our solar system. Large SEP events, although rare, can be very intense. For an overview of the space radiation environment and its 2

ACCEPTED MANUSCRIPT

potential risks to astronauts, the reader is referred to Simpson (1983), Townsend (2005), NCRP (2006), Durante and Cucinotta 2011, Cucinotta et al. (2013), and Straume (2015).

The emphasis here is to provide space-relevant radiation information useful for designing biological experiments in space. To that end, we provide: (a) nuclide-specific dose rates for the GCR spectrum in interplanetary space, (b) particle track traversal rates in

CR IP T

hypothetical biological targets, (c) relationships between particle flux and dose rate, (d) comparisons of radiation environment in interplanetary space with that inside the ISS, (e) implications of particle track microstructure, and (f) probability-dose relationships for solar energetic particle events.

AN US

Nuclide-Specific Dose Rates

Listed in Table 1 are nuclide-specific dose rates vs aluminum shielding thickness calculated for GCR radiation in interplanetary space at 1 astronomical unit (AU). The shielding was a solid sphere of Al with an evaluation point (detector location) at the center of

M

the sphere. Absorbed dose (D) rates and dose-equivalent (DE) rates were calculated using OLTARIS (Singleterry et al. 2011), a validated web-based radiation analysis tool developed

ED

by the NASA Langley Research Center. The D rate was calculated by integrating the product of the flux and LET over all energies. The DE rate was similarly obtained by integrating the product of the flux, LET, and quality factor over all energies. Included in Table 1 are the

PT

resultant average quality factors based on ICRP (2007). These quantities are for 1977 solar minimum GCR conditions (O'Neill 2010) and are scalable for mission duration based on that

CE

environment.

AC

It is observed in Table 1 that some nuclides produce much larger dose rates than

others. This is in part because of the well-known "even/odd" effect (Simpson 1983), i.e., even numbered nuclei are more stable and therefore relatively more abundant. The lightest elements (Z = 1-4) are exceptions. A handful of nuclides in the GCR spectrum contribute most of the dose. With 1 g/cm2 Al shielding, 63% of the D rate is contributed by H and He alone. In contrast, these elements contribute less than 20% of the DE rate behind the same shielding. Four nuclides in the GCR spectrum (Z = 8, 12, 14, and 26) contribute almost half of the DE. Of particular note is that Z = 26 (Fe) contributes almost a quarter of the total DE 3

ACCEPTED MANUSCRIPT

from GCR for 1 g/cm2 Al shielding. The total D and DE rates for the GCR spectrum in interplanetary space (with 1 g/cm2 Al shielding and based on the 1977 solar minimum) was calculated to be 0.38 mGy per day and 2.34 mSv per day, respectively.

It is noted that there is uncertainty in the absolute value of quantities in Table 1 (as well as in subsequent results presented in this paper), originating with the uncertainty in the

CR IP T

particle flux, the data and assumptions used in the modeling, and calculation of D and DE. The total end-to-end model uncertainty (from a known solar cycle to DE) has been estimated to be about 20% (Slaba et al. 2011, 2013, Zeitlin et al. 2013, Matthia et al. 2016). The uncertainty associated with specific biological response functions (which would be highly

AN US

endpoint dependent) is outside the scope of this paper.

It is also observed in Table 1 that the nuclide-specific D and DE contributions are affected by shielding. Most notable, the DE contribution from neutrons (Z=0) becomes significant with thicker shields. Small lightly shielded spacecraft do not result in substantial neutron dose because few neutrons are produced via GCR or SEP primary particle

M

interactions and most of those that are produced escape the spacecraft before they can deposit their energy. This is illustrated in Fig. 1 for GCR and a large SEP event interacting with a

ED

spherical shield with radius equal to the listed shield thickness. For GCR, the dose rate without neutron contribution (solid circles) is plotted as a function of Al shielding thickness and compared with the dose rate from neutrons produced in the shield (open circles). For SEP,

PT

the total dose from the event without neutron contribution (solid circles) is plotted as a function of Al shielding thickness and compared with the dose from neutrons produced in the

CE

shield (open circles).

AC

Effective shielding of small biosentinel spacecraft including their payloads is likely to

be only a few g/cm2 and therefore not expected to experience much neutron dose. However, it should be noted that deep-space vehicles for future human missions and habitats will likely have thick shielding and therefore contributions from neutrons and light ions may dominate the total DE (Norbury and Slaba 2014). Such differences will have to be considered when using results obtained from bio-sentinel missions designed with thin shielding to infer biological responses behind the thicker shields expected for human missions (Curtis and Letaw 1989). 4

ACCEPTED MANUSCRIPT

1.E+00

1.E-02 1.E-03 1.E-04

CR IP T

D rate, mGy/d

1.E-01

1.E-05 0

50

0

50

100

150

1.E+04 1.E+02

AN US

D, mGy

1.E+06

1.E+00 1.E-02

100

150

Depth in Al, g/cm2

PT

ED

M

Figure 1. The contribution of neutrons to dose as a function of Al shielding thickness for GCR and SEP radiation. The GCR (top) is dose rate based on 1977 solar minimum conditions and the SEP (bottom) is total dose for the event based on the uncommonly large August 1972 event. Open circles are for neutrons and closed circles are for dose without neutrons. Calculated using OLTARIS with Al sphere shielding.

CE

Particle Track Traversal Rates in Biological Targets

Table 2 provides the calculated integral flux for each nuclide in GCR and the

AC

calculated traversals per day for individual cell nuclei (additional considerations are required for multiple cells in a sample, as discussed later). For these examples we assumed a spherical 8 µm diameter mammalian cell nucleus, a 2.5 µm diameter spherical yeast cell nucleus, and an 1 µm x 2 µm rod-like bacterium. The calculations assumed 1977 solar minimum conditions (O'Neil 2010) and 1 g/cm2 Al shielding. The nuclide-specific flux can be used to estimate traversal rates for any target size, with considerations for shielding, the nature of the biological sample, and solar cycle effects on GCR.

5

ACCEPTED MANUSCRIPT

It's important to note that in space, particle traversals are essentially random in both time and direction and for small biological targets such as cells and organelles the traversals are rare (Curtis and Letaw1989). As seen in Table 2, the expected GCR particle traversal rates are substantially less than one per day for any of these biological targets. For a typical mammalian nucleus, the traversal rate is about 1 per week from H plus He but less than 1 per year from the heavy ions. For a yeast nucleus it is about one H plus He traversal every two

CR IP T

months, and for a bacterium it is about one per 5 or 6 months.

GCR particles are mostly high energy and will pass through the entire biological sample in a straight line. For a three-dimensional multicellular tissue sample or cells in suspension, this could result in many cells (nuclei) in the sample being traversed by each

AN US

track. It is also the case that the radiations in space are omni-directional, which could potentially result in two or more tracks passing through the same cell but from different directions. As a result, the data in Table 2, while correct, must be used with care. These values represent the averages for individual cells in a tissue sample that is large compared to the distance between individual GCR tracks. However, when one cell nucleus is hit there is a

M

chance that additional cell nuclei along the path of the charged particle will also be hit at essentially the same time. The number of nuclei hit depends on the spacing between nuclei

ED

and the characteristics of the cosmic ray particle. This is discussed in more detail under

AC

CE

PT

section "Track Structure Considerations".

6

ACCEPTED MANUSCRIPT

Yeast nucleusc 1.56E-02 1.64E-03 6.09E-06 3.87E-06 1.25E-05 4.37E-05 1.18E-05 4.13E-05 8.28E-07 6.29E-06 1.33E-06 8.11E-06 1.38E-06 6.06E-06 2.42E-07 1.17E-06 2.43E-07 4.63E-07 3.48E-07 8.80E-07 1.80E-07 6.03E-07 2.92E-07 5.91E-07 3.83E-07 4.04E-06 2.01E-08 1.89E-07

Bacteriumd 6.37E-03 6.70E-04 2.48E-06 1.58E-06 5.08E-06 1.78E-05 4.80E-06 1.68E-05 3.38E-07 2.56E-06 5.41E-07 3.31E-06 5.61E-07 2.47E-06 9.85E-08 4.76E-07 9.90E-08 1.89E-07 1.42E-07 3.59E-07 7.34E-08 2.46E-07 1.19E-07 2.41E-07 1.56E-07 1.64E-06 8.20E-09 7.72E-08

Table 2. Nuclidespecific flux and mean traversals per day calculated for GCR in inter-planetary space at 1 AU for three hypothetical biological targets.a

M

AN US

CR IP T

Mammalian nucleusb 1.60E-01 1.68E-02 6.24E-05 3.96E-05 1.27E-04 4.48E-04 1.21E-04 4.23E-04 8.48E-06 6.44E-05 1.36E-05 8.30E-05 1.41E-05 6.21E-05 2.47E-06 1.20E-05 2.49E-06 4.74E-06 3.56E-06 9.01E-06 1.84E-06 6.17E-06 2.99E-06 6.06E-06 3.92E-06 4.13E-05 2.06E-07 1.94E-06

ED

Integral Flux per cm2-day 3.19E+05 3.35E+04 1.24E+02 7.88E+01 2.54E+02 8.91E+02 2.40E+02 8.42E+02 1.69E+01 1.28E+02 2.71E+01 1.65E+02 2.81E+01 1.24E+02 4.92E+00 2.38E+01 4.95E+00 9.44E+00 7.08E+00 1.79E+01 3.67E+00 1.23E+01 5.96E+00 1.21E+01 7.81E+00 8.23E+01 4.10E-01 3.86E+00

AC

CE

PT

Nuclide Z 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

a

For 1 g/cm2 Al shielding. Calculations using OLTARIS are based on 1977 solar minimum environment (O'Neil 2010). For single cells only. Sample-dependent adjustments to these calculations are required for multicellular samples. 7

ACCEPTED MANUSCRIPT

b

8 µm diameter. 2.5 µm diameter. d 1 x 2 µm rod like bacterium. c

The Relationship Between Nuclide-Specific Flux, D, and DE

CR IP T

It is observed in Fig. 2a that the GCR particle flux is strongly weighted toward the lighter elements, primarily H and He. At the high velocities characteristic of GCR these particles have stopping powers in the same range as the secondary electrons produced by x and γ rays (NIST 2017). These low LET values persist until the ions slow to a small fraction of their original velocity. This results in most of the hits in a biological sample being from

AC

CE

PT

ED

M

AN US

low Z (and relatively low LET) nuclides.

8

ACCEPTED MANUSCRIPT

(a)

1.E+05 1.E+04 1.E+03 1.E+02 1.E+01 1.E+00

(b) 1.E-01

1.E-02

AN US

Dose rate (mGy/d)

1.E-01 1.E+00

1.E-03

1.E-04 1.E+00

(c)

M

1.E-01

ED

1.E-02

1.E-03

1.E-04

PT

Dose-equivalent rate (mSv/d)

CR IP T

Flux (par*cles/cm2-d)

1.E+06

0

5

10

15

20

25

30

CE

Element Number (Z)

AC

Figure 2. (a) Nuclide-specific flux for GCR spectrum. (b) Nuclide-specific absorbed dose (D) rates for GCR spectrum. (c) Nuclide-specific dose-equivalent (DE) rates for GCR spectrum. All were calculated using OLTARIS for 1977 solar minimum conditions and behind 1 g/cm2 Al shielding.

In contrast to the flux in Fig. 2a, it is seen in Figs. 2b and 2c that the higher Z elements

in GCR contribute relatively more to the D and DE than the lighter elements for thin shielding. This is particularly the case for DE (Fig. 2c). In that case, the highest DE rate is not from H and He but rather from Z=26 (Fe). The high Z elements have larger associated Q, which substantially increases DE. Also, the time in the solar cycle can further modify the relative contribution of heavy ions to the fluence and dose, although this is a much smaller 9

ACCEPTED MANUSCRIPT

effect than contributed by increased Q. Hence, both shielding and solar cycle can have implications for biological experiments in deep space, especially for biological endpoints with large relative biological effectiveness (RBE), which can be identified using ground-based simulated space radiations. The above is for thin shielding (1 g/cm2 Al). As shielding is increased (Fig. 3), the

CR IP T

heavier elements are impacted relatively more than the lighter elements. For example, Z=26 (56Fe) produces a larger DE than either H or He with a 1 g/cm2 shield, but much smaller DE when shielding thickness is increased to 30 g/cm2. This effect is mainly due to nuclear interactions resulting in fragmentation of the high Z nuclides into a spectrum of lighter particles and nucleons. In fact, one can see from Fig. 3 that as shielding is increased, the DE

AN US

for H and He actually increases, reflecting gains in their flux from the break-up of heavier ions. Again, this can be an important consideration depending on spacecraft design and

1 g/cm2 Al 10 g/cm2 Al 30 g/cm2 Al

M

1.E+00

ED

1.E-01

PT

1.E-02 1.E-03

CE

Dose-equivalent rate (mSv/d)

placement of the bio-experiment within the spacecraft.

1.E-04

AC

0

10

20

30

Element Number (Z)

Figure 3. Influence of shielding on nuclide-specific dose-equivalent (DE) rates for GCR spectrum. Calculated using OLTARIS for solar minimum conditions.

Interplanetary Space vs the International Space Station (ISS) A major purpose for the ISS is to use it as a space research platform. It is therefore instructive to compare the radiation environment inside the ISS with that in interplanetary 10

ACCEPTED MANUSCRIPT

space at 1AU. Dosimetric quantities are compared in Table 3. Two observations are most notable: (1) the D rate is higher in interplanetary space than inside the ISS even for comparable shielding, and (2) the quality factor (Q) is substantially larger in space with light shielding than inside the ISS resulting in a larger DE rate. These two differences converge to make it more likely that radiation-induced biological responses may be detected on a lightly

CR IP T

shielded biosentinel mission in interplanetary space than on the ISS.

Table 3. Comparison of dosimetric quantities inside the ISS and in interplanetary space at 1AU with various thicknesses of Al shielding (calculated using OLTARIS).a Location D (mGy/d) DE (mSv/d) Q 0.50

Space (1g/cm2)

0.38

2.34

2

Space (10g/cm )

0.38

Space (30g/cm2)

0.36

a

2.3 6.1

AN US

0.22

ISS (service module)b

1.65

4.3

1.10

3.1

M

These dose rates are more than 50 times greater than from natural background radiation on Earth (NCRP 1987, NRC 1990). b Nominal shielding is about 20 g/cm2 Al.

ED

To illustrate this point, we selected an established radiobiological model system (reciprocal chromosome translocations in human lymphocytes) to explore the potential impact of parameter uncertainties on the ability to detect a response from biological experiments in

PT

space. Chromosome translocations in human lymphocytes were selected for this illustration because they have been extensively characterized radiobiologically (both in vitro and in vivo),

CE

and have been used in retrospective biodosimetry on Earth (e.g., Lucas et al. 1992a, 1992b, Straume and Lucas 1995, Lucas et al. 1996, Straume and Bender 1997) as well as in LEO

AC

using pre- and post-mission samples (e.g., Yang et al. 1997, George et al. 2001, 2005, Durante 2005). However, essentially no long-duration testing of chromosome translocation systems (animals or cells) has been performed in space that may be suitable for biosentinel-type missions.

Here, we use this model system for the limited purpose of comparing hypothetical radiation-induced responses on the ISS vs. in a lightly shielded biosentinel beyond LEO (Fig. 4). Designing an actual cytogenetic experiment for space is beyond the scope of this paper. 11

ACCEPTED MANUSCRIPT

For the low-level chronic radiation exposures from GCR in space (whether on the ISS or beyond LEO) it is assumed that the frequency of chromosome translocations is fully stable with time post exposure resulting in a translocation frequency that increases linearly with accumulated DE according to Eqn. 1. It is noted that this assumption has not yet been verified for any biological system in space. In fact, one may expect that heavy ions in GCR would introduce instability in the translocation frequency via the production of complex

CR IP T

rearrangements. Also, it is possible that the heavy shielding of the ISS may influence the RBE differently than a lightly shielded biosentinel such that the calculated DE may not fully track the relative responses in the two locations (e.g., Durante et al. 2005). For our purposes here, however, it is assumed that Eqn. 1 would be representative of any biological system that fully integrates the radiation-induced damage during the mission and that RBE correlates with

AN US

LET. As there are many yet unknown factors contributing to the uncertainty in biological response, a range of overall uncertainties have been used in this illustration (see below and Fig. 4).

Eqn. 1

M

Translocations per cell = b + DE,

where b is the background frequency prior to the mission,  is the linear coefficient obtained

ED

from ground-based calibration measurements, and DE is the total integrated dose equivalent from space radiation during the mission. For the purpose of this illustration, b is 0.005 translocations per cell obtained from "unexposed" subjects in their 30's and 40's (Straume and

PT

Lucas 1995),  is 0.03 translocations per cell per Gy obtained for chronic 137Cs gamma rays (Lucas et al. 1997), and DE is from Table 3 times days in space. Note that using DE from

CE

Table 3 in Eqn. 1 converts  for 137Cs gamma rays to  for ISS or to  for a lightly shielded biosentinel based on the average Q obtained from OLTARIS (it is understood that DE is not

AC

identical to RBE but this approach is considered appropriate for this illustration).

Based on the above model system, Fig. 4 provides an illustration of the potential

impact of uncertainties on the ability to resolve radiation-induced biological effects in space. The uncertainty in the DE is taken as ±20%. The uncertainty in the biology will depend on the specific biological system and endpoint hence we have assumed a range of uncertainties in biological response for this illustration. Radiobiological experiments in well controlled ground-based facilities can easily be ±30% so estimates in the ±40% or even ±50% range 12

ACCEPTED MANUSCRIPT

would not be considered unreasonable for such complex, remote experiments.

It is apparent from Fig. 4 that measuring statistically significant responses above preflight background would be difficult on the ISS. It would require a highly quantitative and reproducible biological system that integrates radiation damage. However, the problem with systems that integrate radiation damage, i.e., damage is stable once produced, is that they

CR IP T

would also have high background. The development of biological systems to measure radiation responses on the ISS, or in other heavily shielded vehicles in LEO, would therefore be challenging.

In contrast, the illustration in Fig. 4 suggests that lightly shielded biosentinel missions

AN US

beyond LEO may improve chances for the detection of significant responses to GCR if

AC

CE

PT

ED

M

suitable biological and technological flight systems were available.

13

ACCEPTED MANUSCRIPT

0.06

ISS

Space

0.05

(a)

0.04 0.03

CR IP T

0.01 0.00 0.06 0.05

(b) 0.04

AN US

0.03 0.02 0.01 0.00 0.06

(c)

0.04 0.03

0.00

PT

0.02 0.01

M

0.05

ED

Es mated transloca ons per cell based on Eqn. 1

0.02

100

CE

0

200

300

400

500

600

700

Time in space (days)

AC

Figure 4. Illustration of biological response as a function of time on the ISS (solid circles) or beyond LEO with light shielding (open circles). Assumed uncertainties: (a) 20% on dose and 30% on response. (b) 20% on dose and 40% on response. (c) 20% on dose and 50% on response.

The effect of shielding is further illustrated in Fig. 5 where nuclide-specific flux inside the ISS is compared with that in interplanetary space for various thicknesses of Al shielding. The median shielding inside the ISS service module is about 20 g/cm2 Al and the shield thickness for a deep-space human mission will likely range from 20 g/cm2 to 40 g/cm2. 14

ACCEPTED MANUSCRIPT

1.E+06 Space GCR (1 g/cm2 Al) Space GCR (10 g/cm2 Al) Space GCR (30 g/cm2 Al) 1.E+04

Inside ISS

1.E+03

CR IP T

Integral Flux (per cm2-day)

1.E+05

1.E+02 1.E+01 1.E+00

AN US

1.E-01 1.E-02 0

5

10

15

20

25

30

Element Number (Z)

ED

M

Figure 5. Comparison of nuclide-specific flux inside the ISS service module with that calculated for interplanetary space at 1 AU with 1, 10, and 30 g/cm2 Al shielding. Based on modeling calculations using OLTARIS and 1977 solar minimum environment (O'Neil 2010).

It is noted in Fig. 5 that as shielding is increased in interplanetary space to thicknesses comparable to the ISS the differences between the nuclide-specific particle flux inside a deep

PT

space vehicle and inside the ISS are reduced, but not eliminated. This difference can be attributed to the geomagnetic field and terrestrial blockage, which attenuate ISS exposures

CE

compared to free space. Behind only 1 g/cm2 of Al shielding, integral flux values for heavy ions (Z > 2) were noticeably larger, as would be expected, and suggests that biosentinel

AC

missions will encounter a radiation environment with a larger concentration of high LET particles compared to what astronauts might encounter in deep space or in the ISS. This must be taken into account when using information obtained from lightly shielded experiments to estimate risk associated with heavier shielded deep-space human missions.

15

ACCEPTED MANUSCRIPT

Track Structure Considerations

The number of track traversals and energy deposited per traversal are not the only considerations for radiation damage in the biological sample. The ionizing particle track itself must be considered. The track can be thought of as composed of a core of dense ionizations and a penumbra of delta rays (electrons) emitted radially from the core. The core of a high Z

CR IP T

particle track produces very dense damage (high LET) and for heavy ions is often lethal to the cell if traversing the nucleus. The penumbra is less dense (consisting of lower LET electron tracks) and produces a dose distribution that decreases with the square of the distance from the core center (Chatterjee and Schaefer 1976, Chatterjee and Holley 1993). There is no sharp distinction between the core and penumbra or cutoff at the limit of the penumbra. The values

AN US

given by the Chatterjee model are intended to represent regions relevant to radiation chemistry processes.

An illustration of the sizes of the track core and penumbra as a function of particle kinetic energy is provided in Fig. 6. It is seen that the core radius is small

M

compared to dimensions of biological cells and organelles and will therefore require a rather precise hit to damage a target via direct action. The radius of the core is less than about 0.01

ED

µm for essentially all GCR and solar particle radiations. In contrast, the penumbra (delta rays) emanating from the core is much larger and can extend to more than 1 mm for very high-energy particles, and therefore can deposit energy in biological targets at substantial

AC

CE

PT

distances away from the central particle track.

16

ACCEPTED MANUSCRIPT

Core

Penumbra

1.E+04

1.E+02 1.E+01 1.E+00 1.E-01 1.E-02 1.E-03 0

500

1000

1500

CR IP T

Radius, µm

1.E+03

2000

2500

AN US

Particle energy, MeV/n

Figure 6. Core radius and penumbra (delta-ray) radius as a function of particle energy. Plotted from data provided in Chatterjee and Schaefer 1976, which are empirically derived based on particle tracks in an aqueous medium. A subsequent derivation of the radius of the core and penumbra was presented in

M

Chatterjee and Holley (1993), and provides similar results, albeit some differences are apparent in the penumbra radius at the highest particle energies. However, as previously

ED

mentioned, there is no sharp cutoff at the limit of the penumbra and the differences in the

PT

derivations do not have practical implications for this paper.

Dose in the track core (Dc) as well as from delta rays (Dp) at various distances from

CE

the track center can be calculated for single particle tracks using Eqns. 2 and 3 below obtained from Metting et al. 1988. Eqn. 2

Dp = 16 (L/2)/(2 b2 ln(2.718*rp/rc)

Eqn. 3

AC

Dc = 16 (L/2)/ rc2 + (L/2)/2 rc2 ln(2.718*rp/rc)

where D is in cGy, L is the unrestricted LET in keV/µm, rc and rp are the radius of the core and penumbra in µm, and b is the radial distance from the center of the core in µm.

17

ACCEPTED MANUSCRIPT

In the above equations, inserting rc and rp calculated by the method provided in Chatterjee and Schaefer (1976) for 250 MeV protons, 200 MeV/n 12C, and 1 GeV/n 56Fe, and the corresponding L for these particles, provides the results in Fig. 7. It is observed that the mean (expected) dose in the track core is high for all particles and that the dose from deltarays diminishes sharply with distance from the core. From these results one may infer that to receive a dose from delta-rays large enough to provide a detectable response in a biological

CR IP T

system would require the "target" to be in close proximity to the track core. For example, the expected delta-ray dose from a 1 GeV/n 56Fe particle is about 16 cGy at 1 µm and only about 1 cGy at 4 µm, which would require a very sensitive biological endpoint to detect. It has been suggested that beyond 4 µm the targets would be hit by only single electron events, which are

AN US

unlikely to result in a detectable biological response (Curtis 2012).

1.E+07

Penumbra - Protons Penumbra - Carbon

1.E+05

Penumbra - Iron

1.E+04

Core - P

1.E+03

Core - C

M

1.E+02 1.E+01 1.E+00

Core - Fe

ED

Dose per par cle track, cGy

1.E+06

1.E-01 1.E-02

PT

1.E-03 1.E-04

CE

0

2

4

6

8

10

12

Distance from center of track, µm

AC

Figure 7. Estimates of average dose (expected value) in single particle tracks from 250 MeV protons, 200 MeV/n 12C, and 1 GeV/n 56Fe. Doses are provided for the track core itself and for the penumbra at various distances from the core.

It should be noted that both the track core and the penumbra have fine structure

(Chatterjee and Holley 1993), which can impact the biological response. The core contains "spurs" that depend on Z and velocity. For light ions such as H and He, the spurs are widely spaced along the length of the core, and for heavy ions, the spurs overlap forming a continuous column. In the penumbra, the delta-ray electrons are separated from each other by 18

ACCEPTED MANUSCRIPT

relatively large distances along the primary track, which results in a small average dose from delta rays because a large fraction of the cells have no energy deposition. However, if a small sensitive target (e.g., DNA) is hit by a delta-ray electron it can deposit much more energy than inferred from the average absorbed dose (Braby 2008). Evaluating such fine structure would require Monte Carlo modeling, which is beyond the scope of this paper.

CR IP T

Biological samples employed in biosentinel experiments can be of various forms, e.g., cells in dry state, cells in suspension, cells in a monolayer adhering to a surface, or

multicellular tissue (which could be either a sample or an entire organism). The form of the sample can influence the microdosimetry and therefore the biological response to the radiation environment. Cells that can exist for extended periods of time in the dry state and

AN US

then begin growing only after a liquid medium is added could be particularly useful for extended biosentinel-type missions in deep space. An example of such biological systems would be yeast, which will be employed in an upcoming NASA BioSentinel mission using radiation sensitive genetically-engineered strains (Bhattacharya et al. 2016).

M

For a sparse monolayer of cells on a flat surface, the likelihood of a track passing through two or more cells is very small. Under such conditions, it is observed in Fig. 8 that

ED

only the two lightest elements (Z = 1-2) are expected to traverse every yeast cell nucleus during missions lasting 3 to 18 months. The expected average hits per nucleus from these elements range from about 1.5 at 3 months to almost 10 at 18 months. The heavier elements

PT

are very unlikely to produce multiple traversals in a nucleus during the mission. This makes it particularly important to consider sample size (the number of cells per sample) when

CE

designing the experiment, i.e., the expected number of cells traversed per sample would be the

AC

probability of traversals per cell times the number of cells per sample.

19

ACCEPTED MANUSCRIPT

1.E+01 Z = 1-2

Z = 11-20

1.E+00

Z = 26 (Fe) Z = 21-28 1.E-01

1.E-02

1.E-03

1.E-04 0

3

6

9

12

15

21

AN US

Time in Space, Months

18

CR IP T

Expected Hits per Yeast Nucleus

Z = 3-10

Figure 8. Particle track traversals per yeast nucleus if single layer on a flat surface as a function of time in interplanetary space. For selected groupings of Z as indicated in the Figure legend. Fig. 9 shows the Poisson distribution (Haight 1967) of Z = 1-2 traversals through a

M

yeast nucleus during 12 months in interplanetary space. The expected value is 6.3 traversals per nucleus. The probability of no traversals is small, 0.2%. This implies that 99.8% of the

ED

nuclei will be traversed by at least one H or He particle during 12 months in space. If the traversals are randomly distributed over time, an average of about one traversal every 2

PT

months would be expected. This is a long time between hits such that one would expect the

AC

CE

GCR environment to produce effects characteristic of low dose-rate exposures.

Figure 9. The Poisson distribution of GCR traversals (Z = 1-2) per yeast nucleus during 12 months in interplanetary space. 20

ACCEPTED MANUSCRIPT

Traversal probabilities were also calculated for the other GCR nuclides during 12 months in space. P (Z=3-10) = expected value is 0.0461, at least one hit is 0.045, no hits is 0.955. P (Z=11-20) = expected value is 0.0074, at least one hit is 0.007, no hits is 0.993. P (Z=21-28) = expected value is 0.0023, at least one hit is 0.002, no hits is 0.998.

CR IP T

P (Z=26) = expected value is 0.0015, at least one hit 0.001, no hits is 0.999. Assuming 105 cells are uniformly distributed in a monolayer on the bottom of a 5 mm diameter well, approximately 160 will be within the penumbra of a 200 MeV/n particle, as described by Chatterjee and Schaefer (1976). Using Eqn. 3, the dose at a given distance from the core can be estimated for tracks perpendicular to the cell monolayer. In this case, the

AN US

"dose" is the average energy per mass for cells hit by delta rays and cells not hit. The energy per mass for single delta ray events is approximately 0.9 cGy (Metting et al 1988) so if, for example, the average dose at a particular distance from the center of the track is 0.1 cGy, approximately 11% of the yeast nuclei at that distance would be hit by a delta ray, i.e., (0.1/0.9) x 100. It is noted that in the dry state there will be only gas or vacuum on the

M

topside of the monolayer hence particles approaching from the top will not produce delta-ray equilibrium at the monolayer and must be considered in the delta-ray dosimetry.

ED

For cells in suspension, if 105 cells are dispersed in 100 µL solution, the cell concentration based on cell volume-to-solution volume would be about 1.1 x 10-4. Although

PT

this would be a dilute cell suspension with substantial average distance between cells, a track (core plus penumbra) passing through the cell suspension would have a significant probability

CE

of passing through more than one cell nucleus. If the cells are randomly positioned in the well, each cell would occupy a neighborhood measuring 100 x100 x 100 µm. Then each track

AC

would on average traverse about 46 of these neighborhoods, i.e., (105)0.33. Because of the very small cross-sectional area of the track core there is only a 0.022 probability that a nucleus (yeast used for this example) will be traversed by a single-track core passing through the suspension of cells described above. For the flux and spectrum of GCR particles in interplanetary space the expected number of track core traversals of nuclei in a sample of 100,000 yeast cells during a 12-month mission would be 6.31x105 (Z = 1-2), 4.61x103 (Z = 310), 7.38x102 (Z = 11-20), and 2.30x102 (Z = 21-28).

21

ACCEPTED MANUSCRIPT

In contrast, the penumbra is much larger than the core and therefore has much greater probability of hitting the nucleus. Listed in Table 4 are the calculated average doses and number of delta-ray hits per nucleus at selected distances from the core center of individual tracks for 1 GeV/n 56Fe, 200 MeV/n 12C, and 250 MeV protons. Again, these are delta-ray hits from a single track passing through the cell suspension described above. For a yeast nucleus, one delta ray hit is estimated to result at about 4 µm from a 1 GeV/n 56Fe track, at about 1.5

CR IP T

µm from 200 MeV/n 12C, and at <0.3 µm from a 250 MeV proton track. At larger distances the probability of a nucleus being hit by a delta ray decreases rapidly. For most biological endpoints a single delta-ray hit (0.9 cGy for a yeast nucleus) may be neglected (Curtis 2012), but for some very radiosensitive endpoints it could potentially be significant.

AN US

The number of delta rays per nucleus during a mission in interplanetary space would be proportional to the integral particle fluence in the biological sample. Detailed Monte Carlo track simulation calculations are needed to produce accurate probabilities for delta ray interactions in cell nuclei at specific distances from GCR tracks. The results in Table 4 are estimates derived from the absorbed dose as a function of distance and measured

M

characteristics of HZE tracks (Brooks et al. 2001, Metting et al. 1988).

ED

Table 4. Expected doses and nuclear hits from delta rays as a function of distance from the center of a particle track.a 1-GeV/n 26Fe

Distance (µm)

1570 174 98 63 16 3.9 1.7 0.98 0.63 0.16 0.025 0.0063 0.0016

AC

CE

0.1 0.3 0.4 0.5 1 2 3 4 5 10 25 50 100

a

Number of delta rays per nucleus 1744 194 109 70 17 4.4 1.9 1.09 0.70 0.17 0.028 0.0070 0.0017

PT

D (cGy)

200 MeV/n 12C

D (cGy)

187 21 12 7.5 1.9 0.47 0.21 0.12 0.075 0.019 0.0030 0.00075 0.00019

Number of delta rays per nucleus 208 23 13 8.3 2.1 0.52 0.23 0.13 0.083 0.021 0.0033 0.00083 0.00021

250-MeV 1H

D (cGy)

4.65 0.52 0.29 0.19 0.046 0.012 0.0052 0.0029 0.00186 0.00047 0.0000743 0.0000186 0.0000046

From a single particle track traversing the center of a yeast nucleus. 22

Number of delta rays per nucleus 5.16 0.57 0.32 0.21 0.052 0.013 0.0057 0.0032 0.0021 0.00052 0.000082 0.000021 0.000005

ACCEPTED MANUSCRIPT

Solar Particle Events and Solar Cycle Considerations

The Sun's activity cycles from low to high with about an 11-y period, which has been observed to vary from 9 y to more than 12 y. The variation in the Sun's activity includes changes in the levels of solar radiation and ejection of solar material and changes in the number of sunspots, flares, and other manifestations. Even with their periodicity, it is not

CR IP T

possible at present to accurately predict future minima and maxima. Hathaway (2016) has recently extrapolated the current solar cycle (Cycle 24) to 2020. Given the uncertainties, his estimate suggests the next solar minimum will occur in the 2019 - 2021 time period, which may be an important consideration for biosentinel-type experiments in the planning phase.

AN US

Due to their stochastic nature, SEP events are generally modeled probabilistically (Xapsos et al. 1999). Fig. 10 illustrates such estimates of the dose expected from SEP events in interplanetary space at 1 AU during solar maximum conditions. The results are for 18 months duration and two shielding thicknesses, 0.3 g/cm2 (1.1 mm) and 1 g/cm2 (3.7 mm) Al.

M

Based on these modeling calculations, there is a 50% chance that a dose of 48 cGy or greater will be received from SEP events during an 18-month mission in deep space with 0.3 g/cm2 (1.1 mm) Al shielding. This is reduced to 15 cGy with 1 g/cm2 (3.7 mm) Al. There is

ED

a 10% chance for a dose of at least 356 cGy and 161 cGy behind 0.3 g/cm2 and 1 g/cm2, respectively. As can be seen in Fig. 9 even larger doses are possible, but unlikely. SEP events

PT

are less likely to occur during solar minimum compared to solar maximum. However, for biosentinel type missions with thin shielding, even low intensity events could present a non-

CE

trivial dose from low energy (<100 MeV) protons and should at least be considered before

AC

final mission architecture is defined.

Importantly, for a lightly shielded spacecraft, the dose from a large SEP event could

substantially exceed that received from GCR radiation and therefore render the GCR contribution to the bio-response difficult to interpret. For example, based on the probabilistic estimates in Fig. 10 for 1 g/cm2 Al shielding, an 18 month mission would result in 50% chance that a dose of 15 cGy would be received from SEP events compared with about 21 cGy during that time period from GCR alone. This could be mitigated by either using biological systems that can be "turned on" when a large SEP event is detected (e.g., 23

ACCEPTED MANUSCRIPT

Biosentinel on the EM-1 mission will have an onboard radiation sensor signaling the addition of media to a sample array) or by shielding designs to reduce the dose contribution from SEP events, while not attenuating GCR significantly. Hence, both biological timing and differential shielding within the payload could possibly be used as strategies to obtain bioresponse information for both GCR and SEP radiation.

1 g/cm2 Al

1000

100

10 0

10

20

AN US

Dose (cGy)

0.3 g/cm2 Al

CR IP T

10000

30

40

50

Probability (%)

PT

Conclusions

ED

M

Figure 10. Probabilistic estimates of dose from SEP events in interplanetary space at 1 AU during solar maximum conditions. Based on Xapsos et al. 1999.

Cosmic-ray interaction data are provided to help in the design of biological experiments in space. Nuclide-specific flux and dose rates were calculated using OLTARIS

CE

and these results were used to determine particle traversal rates and doses in selected biological targets. The lightest elements (H and He) have by far the highest flux and therefore

AC

traversal rates in biological targets, but heavy nuclides contribute substantially to the mean dose and dose equivalent for thin shielding and can produce large energy deposition in small biological targets such as cell nuclei. In interplanetary space, the total GCR spectrum would contribute less than 1 particle traversal per week to a cell nucleus, and less than one per year from the heaviest ions. Such rare (low dose rate) events must be considered in the design of biological experiments.

A comparison is provided between GCR in interplanetary space and inside the ISS. 24

ACCEPTED MANUSCRIPT

Although the radiation dose rate inside the ISS is much higher than on Earth, it is below or approaching the lower limit of detection for most biological endpoints and has been a challenge for providing statistically significant radiobiological data when comparing with "unexposed" controls. Beyond LEO, the combination of higher dose rate and larger Q (i.e., relatively more high-LET particles which usually results in larger biological response per unit dose) increases the chances that biological effects may be detected, particularly for lightly

CR IP T

shielded spacecrafts. This could be a justification for biosentinel-type experiments beyond LEO.

Large solar particle events could substantially impact biological experiments beyond LEO, especially for lightly shielded spacecraft. Large events are rare so they may not occur

AN US

during a particular mission, but planning for them is prudent and could possibly be aided by incorporating "biological timing" or payload shielding designs, i.e., SEPs are much easier to shield against than GCR.

Because of the many challenges associated with the success of such missions,

M

extensive ground-based studies would be required to characterize a biological system prior to flight to determine that the response endpoints selected may be detected during the mission.

ED

This would include exposure to simulated space radiation available at accelerator facilities such as NSRL (La Tessa, et al. 2016). A major concern would be the sensitivity of the biological system to the low-dose-rate GCR radiation environment in space, which is at or

PT

near the detection limit for most biological endpoints. Another concern is the lack of facilities that can deliver simulated GCR radiation at dose rates relevant to those in space, requiring

CE

extrapolation over several orders of magnitude introducing additional uncertainties.

AC

To extrapolate biological response from the high dose rates obtainable at ground-based

accelerators to the much lower dose rates in space the biological system should be evaluated by testing with a combination of split-dose methods at accelerators and supplemented with radiobiological characterization of the system using a range of dose rates and LETs, including space-relevant dose rates which can be obtained from gamma ray and neutron sources. It is well known from radiobiology that lowering the dose rate of acute high dose exposures from low-LET radiations decreases the biological effect for the same total dose (NRC 1990). It is also known that a limiting dose rate is reached below which further reduction in dose rate 25

ACCEPTED MANUSCRIPT

does not appear to influence the response per unit dose. In the low dose-rate environment of space, whether in LEO or beyond, it is expected that the dose response would be independent of dose rate because multiple energy deposition events are not expected to be close enough in time and location to interact prior to repair (NCRP 2000). For low LET radiations, this limiting dose rate has been observed at about 200 mGy/day for life shortening in mice (Sacher and Grahn 1964) and at about 300 mGy/day for chromosome aberrations in human

CR IP T

lymphocytes (Lucas et al. 1997), and is expected to extend to higher dose rates for high LET GCR particles (NCRP 2006). Consistent with these observations, the ICRP (1991) suggested a limiting dose rate for low LET radiations of 144 mGy/day. In interplanetary space, the GCR dose rate varies between 0.14 mGy/day and 0.4 mGy/day during the solar cycle (NCRP 2006). In principle, then, if split-dose experiments at a facility such as NSRL demonstrates

AN US

dose rate independence for GCR particles using space-relevant total dose one may conclude that the response for a particular biological system and endpoint may be linearly extrapolated to the lower dose rates expected in space.

M

Acknowledgments

ED

The work was performed under the auspices of the NASA Ames Research Center, the NASA Langley Research Center, and the Texas A&M University. The work was supported in part by the NASA Advanced Exploration Systems (AES) Division to SB and the NASA

PT

Human Research Program (HRP) to TCS and TS. We also thank Tony Ricco for helpful

CE

comments on the draft paper. The authors have no conflicts of interest concerning this paper.

AC

References

Bhattacharya, S., Ricco, A.J, Hanel, R.P. and Crusan, J. (2016) http://www.nasa.gov/centers/ames/engineering/projects/biosentinel.html (accessed January 9, 2017). Braby, L.A. (2008) A solution to the problem with absorbed dose. Nucl. Eng. Tech. 40:533538. Brooks, A., Bao, S., Rithidech, K., Couch, L.A. and Braby, L.A. (2001) Relative effectiveness of HZE iron-56 particles for the induction of cytogenetic damage in vivo. Radiat. Res. 155:353–359. 26

ACCEPTED MANUSCRIPT

Chatterjee, A. and Schaefer, H.J. (1976) Microdosimetric structure of heavy ion tracks in tissue. Rad. Environ. Biophys. 13:215-227. Chatterjee, A. and Holley, W.R. (1993) Computer simulation of initial events in the biochemical mechanisms of DNA damage. Advances in Radiation Biology 17:181-226.

CR IP T

Cucinotta, F.A., Kim, M.Y., Chappell, L.J., and Huff, J.L. (2013) How safe is safe enough? Radiation risk for a human mission to Mars. PloS ONE 8(10): e74988. doi:10.1371/journal.pone.0074988 Curtis, S.B. and Letaw, J.R. (1989) Galactic cosmic rays and cell-hit frequencies outside the magnetosphere. Adv. Space Res. 9:293-298.

AN US

Curtis, S.B. (2012) HZE fluence rates and energy depositions from delta rays: how important are they? 23rd Annual NASA Space Radiation Investigator's Workshop, Durham, North Carolina, 2012. Durante, M., George, K., Gialanella, G., Grossi, G., La Tessa, C., Manti, L., Miller, J., Pugliese, M., Scampoli, P. and Cucinotta, F. A. (2005) Cytogenetic Effects of High-Energy Iron Ions: Dependence on Shielding Thickness and Material. Radiat. Res. 164:571–576. Durante, M. (2005) Biomarkers of space radiation risk. Radiat. Res. 164:467-473.

M

Durante, M. and Cucinotta, F.A. (2011) Physical basis of radiation protection in space travel. Reviews of Modern Physics 83:1245-1281.

ED

George K, Durante M, Wu H, Willingham V, Badhwar G, Cucinotta FA (2001) Chromosome aberrations in the blood lymphocytes of astronauts after space flight. Radiat. Res. 156:731738.

PT

George K, Willingham V, Cucinotta FA (2005) Stability of chromosome aberrations in the blood lymphocytes of astronauts measured after space flight by FISH chromosome painting. Radiat. Res. 164:474-480.

CE

Haight, F.A. (1967) Handbook of the Poisson Distribution. New York: John Wiley & Sons.

AC

Hathaway, D. (2016) http://solarscience.msfc.nasa.gov/predict.shtml (accessed February 1, 2017). ICRP (1991) Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. Pergamon Press, Oxford. ICRP (2007) The 2007 recommendations of the International Commission on Radiological Protection. ICRP publication 103, Pergamon Press, Oxford. La Tessa, C., Sivertz, M., Chang, I., Lowenstein, D., and Rusek, A. (2016) Overview of the NASA space radiation laboratory. Life Sci. Space Res. 11:18-23. 27

ACCEPTED MANUSCRIPT

Lucas JN, Awa A, Straume T, Poggensee M, Kodama Y, Nakano M, Ohtaki K, Weier HU, Pinkel D, Gray J, Littlefield G (1992a) Rapid Translocation Frequency Analysis in Humans Decades After Exposure to Ionizing Radiation. Int. J. Radiat. Biol. 62:53-63. Lucas, J.N., Poggensee, M., Straume, T. (1992b) The Persistence of Chromosome Translocations in a Radiation Worker Accidentally Exposed to Tritium. Cytogenet. Cell Genet. 60, 255-256.

CR IP T

Lucas JN, Hill FS, Burk CE, Cox AB, Straume T (1996) Stability of the Translocation Frequency Following Whole-Body Irradiation Measured in Rhesus Monkeys. Int. J. Radiat. Biol. 70:309-318. Lucas, J.N., F.S. Hill C.E. Burk A.D. Lewis A.K. Lucas A.M. Chen F.C. Sales T. Straume (1997) Dose-response curve for chromosome translocations induced by low dose rate 137Cs gamma rays. Radiat. Prot. Dosim. 71:279-282.

AN US

Matthia, D., Ehresmann, B., Lohf, H., Kohler, J., Zeitlin, C., Appel, J., Sato, T., Slaba, T.C., Martin, C., Berger, T., Boehm, E., Boettcher, S., Brinza, D.E., Burmeister, S., Guo, J., Hassler, D.M., Posner, A., Rafkin, S.C.R., Reitz, G., Wilson, J.W., Wimmer-Schweingruber, R.F., The Martian surface radiation environment – a comparison of models and MSL/RAD measurements. J. Space Weather Spac. 6: A13; 2016.

M

Metting, N.F., Rossi, H.H., Braby, L.A., Kliauga, P.J., Howard, J., Zaider, M., Schimmerling, W., Wong, M. and Rapkin, M. (1988) Microdosimetry near the trajectory of high-energy heavy ions. Radiat. Res. 116:183–195.

ED

NASA (2016). https://www.nasa.gov/content/journey-to-mars-overview (accessed February 2, 2017).

PT

NCRP (1987) Exposure of the population in the United States and Canada from natural background radiation. NCRP report 94, Bethesda, MD. NCRP (2000) National Council on Radiation Protection and Measurements, Radiation Protection Guidance for Activities in Low-Earth Orbit. NCRP Report 132, Bethesda, MD.

CE

NCRP (2006) Information needed to make radiation protection recommendations for space missions beyond low-earth orbit. NCRP report 153, Bethesda, MD.

AC

Norbury, J.W. and Slaba, T. (2014) Space radiation accelerator experiments - the role of neutrons and light ions. Life Sci. Space Res. 3:90-94. NIST (2017) Stopping Power and Range Tables for Electrons, Protons, and Helium Ions. Berger, M.J., Coursey, J.S., Zucker, M.A., and Chang, J. National Institute of Standards and Technology, Physical Measurement Laboratory. https://www.nist.gov/pml/stopping-power-range-tables-electrons-protons-andhelium-ions (accessed April 10, 2017). NRC (1990) Health effects of exposure to low levels of ionizing radiation. BEIR V Report, National Research Council, National Academy Press, Washington DC. 28

ACCEPTED MANUSCRIPT

O'Neill, P.M. (2010) Badhwar-O'Neill galactic cosmic ray flux model – Revised. IEEE Trans. Nuc. Sci. 57:3148–3153. Sacher, G.A. and Grahn, D. (1964) Survival of mice under duration-of-life exposure to gamma rays, I. The dosage-survival relation and the lethality function. J. Natl. Cancer Inst. 32:277–321.

CR IP T

Simpson, J. A. (1983) Elemental and isotopic composition of the galactic cosmic rays. Ann. Rev. Nucl. Part. Sci. 33:323-381. Singleterry, R.C., Blattnig, S.R., Clowdsley, M.S., Qualls, G.D., Sandridge, C.A., Simonsen, L.C., Slaba, T.C., Walker, S.A., Badavi, F.F., Spangler, J.L., Aumann, A.R., Zapp, E.N., Rutledge, R.D., Lee, K.T., Norman, R.B. and Norbury, J.W. (2011) OLTARIS: On-line tool for the assessment of radiation in space. Acta Astronautica 68:1086-1097.

AN US

Slaba, T.C., Blattnig, S.R., Badavi, F.F., Stoffle, N.N., Rutledge, R.D., Lee, K.T., Zapp, E.N., Dachev, Ts. P., Tomov, B.T., Statistical Validation of HZETRN as a Function of Vertical Cutoff Rigidity using ISS Measurements. Adv. Space Res. 47: 600-610; 2011.

M

Slaba, T.C., Blattnig, S.R., Reddell, B., Bahadori, A., Norman, R.B., Badavi, F.F., Pion and electromagnetic contribution to dose: comparisons of HZETRN to Monte Carlo results and ISS data. Adv. Space Res. 52: 62-78; 2013.

ED

Straume T, Lucas J (1995) Validation Studies for Monitoring of Workers Using Molecular Cytogenetics," in Biomarkers and Occupational Health: Progress and Perspectives, Mendelsohn ML, Peeters JP, Normandy MJ (eds), Joseph Henry Press, Washington, D.C., pp. 174-193.

PT

Straume T, Bender MA (1997) Issues in cytogenetic biological dosimetry: Emphasis on radiation environments in space. Radiat. Res. 148:S60-S70.

CE

Straume, T. (2015) Medical Concerns with Space Radiation and Radiobiological Effects. In: Handbook of Cosmic Hazards and Planetary Defense (J. Pelton and F. Allahadi, Eds.), Springer International Publishing, Switzerland. DOI 10.1007/978-3-319-03952-7_4.

AC

Todd, P. and J.T. Walker (1984) The Microlesion Concept in HZE Particle Dosimetry. Adv. Space Res. 4:187-194. Townsend, L.W. (2005) Implications of the space radiation environment for human exploration in deep space. Radiat. Prot. Dosim. 115:44-50. Usoskin, I. G., G. A. Bazilevskaya, and G. A. Kovaltsov (2011) Solar modulation parameter for cosmic rays since 1936 reconstructed from ground-based neutron monitors and ionization chambers. J. Geophys. Res., 116 (doi:10.1029/2010JA016105).

29

ACCEPTED MANUSCRIPT

Xapsos, M.A., Barth, J.L., Stassinopoulos, E.G., Burke, E.A., and Gee, G.B. (1999) Space environment effects: Model for emission of solar protons (ESP) – cumulative and worsk case event fluences. NASA Report TP-1999-209763 (December 1999). Yang, T.C., K. George, A. S. Johnson, M. Durante, and B. S. Fedorenko (1997) Biodosimetry results from space flight Mir-18. Radiat. Res. 148:S17-S23.

AC

CE

PT

ED

M

AN US

CR IP T

Zeitlin, C., Hassler, D.M., Cucinotta, F.A., Ehresmann, B., Wimmer-Schweingruber, R.F., Brinza, D.E., Kang, S., Weigle, G., Bottcher, S., Bohm, E., Burmeister, S., Guo, J., Kohler, J., Martin, C., Posner, A., Rafkin, S., Reitz, G., Measurements of energetic particle radiation in transit to Mars on the Mars science laboratory. Science 340: 1080-1084, 2013.

30

ACCEPTED MANUSCRIPT

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

AC

26

CR IP T

8

AN US

3 4 5 6 7

M

2

ED

1

PT

Z 0

CE

Nucli de

Table 1. Nuclide-specific dose rates vs. aluminum shielding for GCR radiation in Average Quality D Rate (µGy/day) DE Rate (µSv/day) Factorb 1 10 30 1 10 30 1 10 30 g/cm2 g/cm2 g/cm2 g/cm2 g/cm2 g/cm2 g/cm2 g/cm2 g/cm2 0.10 1.01 3.24 1.96 20.60 66.65 20.22 20.34 20.55 162.3 239.8 5 216.25 250.70 0 323.21 377.80 1.48 1.49 1.51 222.8 79.58 82.51 69.47 1 269.12 283.95 2.80 3.26 4.09 0.56 0.56 0.44 0.86 0.89 0.66 1.53 1.58 1.51 0.57 0.61 0.52 0.95 1.13 0.95 1.66 1.85 1.84 3.12 2.85 2.02 8.40 7.74 5.08 2.70 2.72 2.51 16.37 11.66 5.97 68.76 44.45 19.77 4.20 3.81 3.31 5.84 4.55 2.60 31.47 23.04 11.77 5.39 5.07 4.53 212.9 27.75 17.47 7.35 2 119.87 44.44 7.67 6.86 6.05 0.65 0.77 0.62 5.41 6.30 4.67 8.35 8.20 7.54 6.46 4.16 1.83 71.26 42.27 16.99 11.04 10.16 9.28 1.59 1.34 0.80 19.66 15.80 8.81 12.35 11.77 10.96 178.5 12.13 6.99 2.61 4 95.74 33.27 14.72 13.70 12.73 2.31 1.57 0.73 36.79 23.89 10.52 15.90 15.21 14.40 216.5 12.19 6.63 2.26 7 112.76 36.59 17.76 17.00 16.19 0.50 0.47 0.29 9.26 8.62 5.17 18.50 18.33 17.84 2.91 1.75 0.68 59.11 35.06 13.34 20.29 20.03 19.62 0.64 0.58 0.33 13.93 12.59 7.10 21.62 21.56 21.36 1.43 0.98 0.43 32.63 22.35 9.78 22.80 22.92 22.93 1.22 0.84 0.38 28.84 20.18 9.16 23.58 24.00 24.32 3.57 1.90 0.64 84.44 46.83 16.21 23.63 24.69 25.44 0.80 0.65 0.33 18.87 16.03 8.51 23.73 24.84 25.90 3.06 1.60 0.55 67.65 37.79 13.56 22.08 23.58 24.73 1.59 1.02 0.42 33.97 23.12 10.01 21.43 22.72 23.76 3.50 1.87 0.63 71.72 40.72 14.32 20.52 21.83 22.84 2.40 1.50 0.57 47.96 31.75 12.54 19.95 21.11 22.02 527.9 27.75 11.48 2.57 8 233.70 54.58 19.03 20.36 21.26 0.15 0.07 0.02 2.71 1.40 0.42 18.53 19.72 20.58 1.45 0.61 0.13 26.21 11.70 2.69 18.12 19.22 19.97 b ainty within ±20%. Based on ICRP 2007.

27 28

31

int erp lan eta ry spa ce at 1 AU. a a

Cal cul ati ons usi ng OL TA RIS wit h 19 77 sol ar mi ni mu m env iro nm ent. Ov era ll unc ert