Beamline and Flight Comparisons of the ARMAS Flight Module With the Tissue Equivalent Proportional Counter for Improving Atmospheric Radiation Monitoring Accuracy

Ionizing radiation at aircraft and commercial suborbital spaceflight altitudes is driven by space weather and is a health concern for crew and passengers. We compare the response functions of two radiation detectors that were exposed to four different ground‐based laboratory radiation fields as well as flown alongside each other on aircraft. The detectors were a tissue equivalent proportional counter (TEPC) and a Teledyne silicon micro dosimeter chip that was integrated into an Automated Radiation Measurements for Aerospace Safety Flight Module (ARMAS FM). Both detectors were flown onboard commercial and research aircraft. In addition, both detectors were exposed neutrons at the Los Alamos Neutron Science Center, protons at Loma Linda University Medical Center, 56Fe particles at the NASA Space Radiation Laboratory, and also a gamma radiation source at Lawrence Livermore National Laboratory. The response of each of these instruments as well as derived dosimetric quantities are compared for each radiation exposure and the ratio for converting ARMAS absorbed dose in silicon to an estimated absorbed dose in tissue is obtained. This process resulted in the first definitive calibration of the silicon‐based detector like ARMAS to TEPC. In particular, with seven flights of both instruments together, the ARMAS‐derived dose in tissue was then validated with the TEPC‐measured dose in tissue and these results are reported. This work provides a method for significantly improving the accuracy of radiation measurements relevant to human tissue safety using a silicon detector that is easy to deploy and can report data in real time.


Introduction
Aerospace safety has improved over the past decades in many areas including pilot training, flight weather specification, air drag efficiency, and avionics redundancy. However, long-term health effects from space weather due to ionizing radiation at higher altitudes and higher magnetic latitudes have not seen the same growth in safety awareness until very recently. Ionizing radiation exposure is a natural hazard faced by commercial aircrew, high-altitude pilots, frequent flyers, first trimester fetuses, and commercial space travelers. In this work we have significantly improved the accuracy of silicon-based radiation detection instrumentation through calibrations of measured dose in silicon converted to dose in tissue. These results will aid our understanding of the impacts from complex radiation fields upon humans.
Multiple sources of ionizing radiation contribute exposure in the aerospace environment from the Earth's surface into space. Galactic cosmic radiation (GCR) and solar energetic particles (SEPs) are the dominant ionizing radiation sources for travel in and above commercial aviation altitudes (Friedberg & Copeland, 2003. A large number of measurements used for postflight analysis have been made using Tissue Equivalent Proportional Counters (TEPCs) under GCR background conditions and not during major space weather SEP events (Beck et al., 1999(Beck et al., , 2005Dyer et al., 1990Dyer et al., , 2009Gersey et al., 2012;Getley et al., 2005Getley et al., , 2010Kyllönen et al., 2001;Latocha et al., 2007;Lindborg et al., 2004;Meier et al., 2009;Tobiska et al., 2014aTobiska et al., , 2014bTobiska et al., , 2015. Some of the measurements have included neutron flux and dose equivalent measurements with solid-state detectors Lee et al., 2015;Ploc et al., 2013).
There are many modeling systems into which ionizing radiation data could be integrated, e.g., LUIN (O'Brien et al., 1996), CARI6PM (Friedberg et al., 1999;Friedberg & Copeland, 2003, FLUKA (Zuccon et al., 2001), QARM (Lei et al., 2006), AIR (Johnston, 2008), PARMA (Sato et al., 2008), AVIDOS (Latocha et al., , 2014, NAIRAS (Mertens et al., 2013), PANDOCA , and KREAM (Hwang et al., 2014). Recent work by Joyce et al. (2014) utilized CRaTER measurements (Schwadron et al., 2012;Spence et al., 2010) in deep space to estimate dose rates through the Earth's atmosphere at a range of different altitudes down to aviation heights. Figure 1 shows an example of that radiation field from the NASA Langley Research Center's (LaRC) Nowcast of Atmospheric Ionizing RAdiation System (NAIRAS) v1 model (Mertens et al., 2013). Geographic latitude and altitude (km) are shown for a western hemisphere longitude at 270°E during NOAA G3 strongly geomagnetically active conditions. The D-index Tobiska et al., 2017) relates the severity of effective dose rates for humans in an atmospheric radiation field. "D" in D-index stands for "dose" and the black, purple, and blue colors indicate lower effective dose rates while green-yellow show moderate effective dose rates and orange-red represent elevated effective dose rates. The D-index was developed to provide warnings of elevated radiation levels. It is based on the radiation exposure from solar particle and radiation belt precipitation added to background GCRs. It is created from the effective dose rate, which can be derived from either measurements or models. The D-index range from D0 to D8 covers a wide range of radiation exposures at aviation altitudes. D0, D1, and D2 levels are for quiet space weather conditions. D3 is for elevated exposure from more particles coming into the atmosphere and can be used by air traffic management to trigger a radiation warning. D4 and higher indicate radiation alerts can occur infrequently but during large solar events.
All longitudes are provided through the public access ARMAS app available in Apple's App store.
GCRs are produced outside the solar system and mostly consist of energetic protons that penetrate through the heliosphere; they are modulated by the strength of the Sun's interplanetary magnetic field (Simpson, 1983). SEPs come from solar activity, i.e., coronal mass ejections related to flaring events or from interplanetary magnetic field shocks (Gopalswamy, 2004;Reames, 2013). A third radiation source has been identified and likely originates from the wave-particle interaction in the Van Allen radiation belts leading to precipitated charged particles (Tobiska et al., , 2018; this is an area of continued active research. The primary ionizing dose in the aerospace environment, which can be altered with both the composition of the terrestrial atmosphere and the dynamic variability of the Earth's magnetosphere, produces highly variable secondary and tertiary ionizing radiations. These particularly cause concern for human health and vehicle avionics (IEC 62396-1, 2016) in higher altitude and higher latitude aviation as illustrated in Figure 1.
Radiation detection in space and aboard aircraft is a challenge due to the complexity of the primary and secondary radiation environment. Ground-based experiments are required to test and benchmark the response of existing and new detectors in order to gain insight into the operational efficacy of their measurements. Two radiation detectors were utilized in this study. We performed a cross-calibration followed by measurement comparisons in four ground-based beamline experiments and seven aircraft flights. The two detectors were a TEPC, the community's "gold standard" for tissue-relevant radiation detection (Badhwar et al., 1992(Badhwar et al., , 1995, and a detector based on the Teledyne micro dosimeter chip called Automated Radiation Measurements for Aerospace Safety Flight Module (ARMAS FM). In this study both of these radiation detectors were flown together onboard aircraft to perform measurements of the ambient radiation field while in flight.

Space Weather
Both detectors were also exposed to four different radiation fields during ground-based experiments. A direct comparison of the absorbed dose, D, measured by each detector during the ground-based experiments and aircraft flights were then made. Further insight was obtained concerning a normalization method for converting dose in silicon, D (Si) (measured by the ARMAS FM), to dose in tissue, D (Ti) (measured by the TEPC), using results from these ground-based experiments. The seven flights were conducted across the midcontinental United States to capture changes in cutoff rigidities at flight altitudes.
The rationale for making the ARMAS-TEPC calibration, with validation, was to develop a system that could give the quality of information provided by TEPC but without the penalties of instrument fragility, heavy weight, and inability to conduct real-time monitoring, as is the case with TEPC. The TEPC data analysis is a complex process and the advantages of the ARMAS FM are superior for robust data streaming its real-time results, low size/weight/power, and ease of operation. Our results improve the accuracy for determining absorbed dose, D (Ti), and dose equivalent, H, using the ARMAS FM detector for nominal exposure conditions. Ground-level events (GLEs) and large solar particle events (SPEs) are not included in this analysis and require modeling as well as measurements from a larger body of work. This accuracy also lays a solid foundation for using silicon-based calibrated, automated detectors to provide a continuous flow of aerospace radiation environment information. The information is particularly useful for monitoring regulatory compliance with guidelines by international organizations such as the International Civil Aviation Organization (ICAO). The most important dosimetric parameters for this monitoring include the total absorbed dose (and rate) in tissue as well as the total dose equivalent (and rate).

Instrumentation and Experimental Facilities
Two radiation detectors were exposed to four different ground-based radiation fields. Their characteristics are described as are the ground-based experiment facilities.

Radiation Detection Instrumentation 2.1.1. TEPC
The TEPC used in the ground-based beamline experiments is an active radiation detector similar in design and identical in function to the TEPC used aboard the space shuttle to provide energy loss spectra for crew dosimetry (Badhwar, 1997;Badhwar et al., 1992;. It has an active volume that is a right circular cylinder 1.78 cm long and 1.78 cm in diameter. The walls of this cylinder are fabricated from tissue equivalent A-150 plastic and are 1.9 mm thick. The cylinder is enclosed in stainless steel walls of 0.89 mm thickness. The active volume is then enclosed in an aluminum cylinder, which measures 2 in. in diameter and is 12 in. long. The active volume of the TEPC is filled with low-pressure tissue equivalent (TE) gas (propane) to simulate tissue material as a right circular cylinder with 2 μm diameter and length (International Commission on Radiation Units and Measurements, 1983). An anode wire runs the length of the cylinder and is kept at a potential of 670 V relative to the cylinder walls.
When energy is deposited in the TEPC active volume, charge is collected at the anode wire and is processed by a preamplifier. The signal then moves from the preamplifier to two shaping amplifiers. These two shaping amplifiers have a difference in gain of approximately a factor of 50. After shaping amplification, the two signals are converted to a digital pulse height by means of an analog to digital converter (ADC). These pulse height signals are then stored on a flash-ROM card for later analysis. Pulse height information as well as instrument health and status are time-stamped and stored to flash-ROM in 1-min increments. Health and status information for the TEPC include reporting of instrument temperature, voltage, and current draw for each minute of operation.
During data analysis, the pulse height signals are combined to form a pulse height distribution. This distribution is then calibrated and converted into a lineal energy (y) deposition distribution (N(y) vs. y). The TEPC is capable of measuring a lineal energy deposition spectrum ranging from 0.4-1,250.0 keV μm −1 . The y spectrum is used to determine the absorbed dose to tissue from the radiation incident upon the TEPC. By making the assumption that y is equal to the linear energy transfer (LET) of the radiation field, this same spectrum is used to approximate the LET spectrum of the incident radiation Badhwar et al., 1995Badhwar et al., , 1996. This LET spectrum is then used to calculate absorbed dose, D, dose equivalent, H, and an average quality factor, Q, of the radiation in tissue (Johnson et al., 1993). Absorbed Dose in Silicon, D (Si), is the amount of energy absorbed by silicon per unit mass. This fundamental radiation measurement 10.1029/2020SW002599 Space Weather quantity for Absorbed Dose in silicon has units of Gray (Gy). Absorbed Dose in Tissue, D (Ti), is the amount of energy absorbed by human tissue per unit mass. This fundamental radiation measurement quantity for Absorbed Dose in tissue has units of Gray (Gy). Dose Equivalent, H, is the radiation quantity used to report a person's exposure for regulatory, medical, and scientific purposes. Regulatory limits are expressed in units of Dose Equivalent. Dose Equivalent is calculated by multiplying the Absorbed Dose (D) times the Average Quality Factor (Q) (H = D × Q). Units for Dose Equivalent are reported in Sieverts (Sv). Average Quality Factor, Q, scales the exposure in a specific radiation field to the potential biological risk. The dose equivalent from the radiation field is determined using the values of the absorbed dose and the average quality factor. Using the time-stamp information, the time profile of dose and dose equivalent is then calculated.
The calculated quantities of absorbed dose, D (Ti), dose equivalent, H, and an average quality factor, Q, of the radiation in tissue are shown in Equation 1. The calculation of absorbed dose (International Commission on Radiation Units and Measurements, 1983), D, in J kg −1 , is where ε is the energy deposited in the TEPC (Joules) as derived from a lineal energy spectrum. N(ε)dε is the number of particle events depositing energy ε. ρ g is the density of the TE gas in the TEPC gas cavity (kg cm −3 ). V g is the volume of the TEPC gas cavity (cm 3 ). The SI unit of absorbed dose is Gray (Gy), which is equal to 1 J kg −1 . We note that the unit for absorbed dose in the United States is called the Rad and 1 Gy is equal to 100 Rad.
The quality factor, Q, is defined as a dimensionless scaling parameter that is a function of the LET of incident radiation into tissue produced by the absorbed dose (International Commission on Radiological Protection, 1991). The dose equivalent, H, is the product of Q and D at a specific point in tissue (International Commission on Radiation Units and Measurements, 1993). This is shown in Equation 2 and is in SI units of Sievert (Sv), where 1 Sv is also equal to 100 REM.
The dose equivalent, H, is used in radiation protection standards for humans and we further note that while Gy is an absorbed dose, Sv is an absorbed dose that takes into account the type of radiation in tissue since some types of radiation are more damaging than others, thus leading to the term "dose equivalent." Q is the scaling factor that transforms absorbed dose to dose equivalent.
The TEPC used for measuring the radiation field onboard aircraft flights, results of which are reported in section 3, is owned by SolarMetrics and flight managed by SET ( Figure 2). It has an active volume that is different in size and shape from the shuttle-style TEPC used in the beamline experiments conducted by Prairie View A&M University (PVAMU) for many years (Gersey et al., 2004;Gersey, Aghara, et al., 2007, Gersey, Sodolak, et al., 2007 but is virtually identical in all other ways including electronics, data management and storage, data analysis. The shuttle-style TEPC has been used in beamline experiments by this team and it has provided reliable and consistent microdosimetry measurements in a multitude of different radiation fields (Gersey, Aghara, et al., 2007;Gersey, Sodolak, et al., 2007;Gersey et al., 2004Gersey et al., , 2012. The TEPC used on aircraft flights is denoted  as a HAWK style and its active volume is a spherical cavity 12.57 cm in diameter with a wall (comprised of A-150 tissue equivalent plastic) 0.241 cm thick; its active volume is filled with low-pressure propane gas to simulate a 2-μm diameter sphere (TEPC Technical Manual, Far West Technologies).

Automated Radiation Measurements for Aerospace Safety Flight Module (ARMAS FM)
The ARMAS radiation measurement flight system uses Teledyne micro dosimeters (uDOS001, Figure 3) developed by The Aerospace Corporation, manufactured under license by Teledyne, and first used by the CRaTER experiment on the Lunar Reconnaissance Orbiter (Mazur et al., 2011;Spence et al., 2010). The uDOS001 directly measures total ionizing dose (TID) absorbed by an internal silicon test mass; this is also called absorbed dose in silicon, D (Si). The entire chip size is 3.56 × 2.54 × 0.10 cm. By accurately measuring the energy absorbed from heavy ions, alphas, protons, neutrons, electrons, and gamma rays (γ rays), an estimate of the absorbed dose in silicon is made. The micro dosimeter provides repeatable measurements of TID in silicon over a wide range of energies (60 keV to >175 MeV) and operating temperatures (−30 to +40°C). In ARMAS, the instrument is typically operated at aircraft cabin temperatures (15-25°C); for environments colder than −15°C, passive thermal insulation is used. The micro dosimeter operates from input power voltages >13 VDC. Figure 4 is a diagram of the ARMAS instrument measurement concept and Figure 5 shows an ARMAS Flight Module 7 (FM7) with Bluetooth capability that is used in business jet and commercial aerospace vehicles.
The accumulated dose is reported via three DC linear and one pseudo-logarithmic output channels in units of millivolts, giving a dose resolution of 14 μrads and a measurement range exceeding 100 krads. Each channel (ch1-ch4) has a resolution of 256 steps, i.e., values of 0-255, and when a lower channel rolls over to 0 the next higher channel is incremented by 1. Level 0 (L0) raw data are reported as the output of the four channels, each having a value of 0-255. In addition to the uDOS001 and ADC providing ch1-ch4 information, a microprocessor (μproc) is used to accumulate the absorbed dose measurements over time; the integration time varies but 10 s is used operationally. Along with an input GPS signal for time and position, the output L0 information is both stored on an internal micro SD card data logger as well as sent externally via one of several systems. Depending upon the configuration of the specific ARMAS instrument, the L0 and GPS information may be sent via Iridium satellite link, Bluetooth to a paired iPhone or iPad via the ARMAS app, Ethernet, USB, RS232, or I 2 C. Measured D (Si) is continuously downloaded from the instrument to the SET ARMAS servers via WiFi or Iridium downlink.

Ground-Based Experimental Facilities
There were four different facilities utilized in this study and each facility provided a different and distinctly well-characterized radiation field. These four radiation fields were chosen to represent important components of the complex mix of radiation found in the atmosphere and in space. The two radiation detectors were exposed to radiation at these facilities under identical conditions so that a meaningful comparison of the measured absorbed dose could be made.

Los Alamos Neutron Science Center (LANSCE) Neutron Source
The LANSCE 30L beamline (Irradiation of Chips Electronics (ICE) House) delivers high energy and relativistic neutrons from 1-800 MeV in energy. The neutron energy spectrum on this beamline is very similar to the secondary neutron energy spectrum found in the upper atmosphere ( Figure 6). One hour of exposure in the neutron beamline is equivalent to 30,000 h at 12 km (39,370 ft.).

Space Weather
The neutron beam at the ICE House is produced by a spallation reaction caused by 800 MeV protons striking a tungsten target. The ICE House beam line is located approximately 20 m from the spallation source and is 30°to the left (30L) of the incident proton beam direction. Both the neutron energy spectrum and the total number of neutrons during each experimental exposure in the ICE House are measured using a fission chamber detector coupled with a Time Of Flight (TOF) analysis system (Wender et al., 1993).
A laser system was used to align both the TEPC and the ARMAS FM detectors in the ICE House neutron beam line as shown in Figure 7. After each detector was exposed to the neutron beam for a given period of time, analysis of the data from each detector was performed resulting in a determination of the total measured absorbed dose. The resulting total absorbed dose for each experimental exposure was then divided by the total number of neutrons per cm 2 for each exposure; this information is provided by the ICE House fission chamber detector system. This normalization procedure allowed a meaningful comparison of results between the two detectors and between the same detector exposed for different lengths of time (Gersey, Aghara, et al., 2007;Gersey et al., 2003).

Loma Linda University Medical Center (LLUMC) Proton Synchrotron
LLUMC is capable of producing high energy protons at energies of interest to the aerospace radiation dosimetry community, e.g., 75-250 MeV. The Horizontal Beam Line (HBL) treatment bay beam line was utilized as the experimental area to expose the two detectors to high energy protons. In this study a 175 MeV proton beam was utilized due to the fact that this energy of protons is prevalent during normal solar activity and during SPEs, especially in the secondary radiation environment. Protons with these energies are also found in the upper atmosphere and can penetrate to internal organs. The TEPC and the ARMAS were exposed to the 175 MeV proton beam separately and the absorbed dose measurements for each detector were determined during post experimental analysis.
The LLUMC synchrotron facility did not have a method to determine the total number of protons delivered to the experimental area during each exposure. Instead, the facility was able to provide a measurement from a foil detection system upstream of the experimental area (denoted Tic1 foil in this study). The number of Tic1 foil counts measured by this system during any given experimental exposure was directly proportional to the number of protons per cm 2 delivered by the synchrotron. The number of Tic1 foil counts measured during each experimental exposure of the TEPC and ARMAS FM was therefore used to normalize the absorbed dose measurements of these devices for comparative purposes.

Space Weather
GERSEY ET AL.

Brookhaven National Laboratory's NASA Space Radiation Laboratory (NSRL) Particle Beam
NSRL is a facility capable of delivering protons and heavy ions at energies found in both the Galactic Cosmic Radiation (GCR) and SEP spectrum in space and at the top of the atmosphere. The species of ions that the NSRL facility is capable of delivering range from H up to 197 Au. The maximum energy that each ion can be delivered at is species dependent and range from 400-2,500 MeV/nucleon. At NSRL the energy and species of radiation chosen for this study was 350 MeV/nucleon 56 Fe ions. This energy and species of particle was chosen to represent the High Z and Energy (HZE) portion of the GCR and SEP spectrum. The absorbed dose measured by each detector after exposure to the 56 Fe beam was normalized by the number of 56 Fe ions per cm 2 . The number of 56 Fe ions per cm 2 delivered during each exposure was determined using a scintillation counter positioned in the particle beam next to the radiation detectors.

Lawrence Livermore National Laboratory (LLNL) Gamma
Ray Source LLNL provided a 60 Co 1.25 MeV NIST traceable gamma ray source for this study. These energetic photons are a component of the radiation field in the upper atmosphere. The ARMAS and the TEPC were placed in a low scatter facility 200 cm away from the 60 Co source during exposure. Further details of this LLNL irradiation facility may be found in Straume et al. (2016). Once the measured absorbed dose was determined for the ARMAS and the TEPC, the results were normalized by the elapsed time for each exposure.

Instrumentation Protocol for Aircraft Flight Exposures
The detectors used for in-flight dosimetry comparisons were simultaneously flown aboard both commercial and research aircraft. In the research vehicle, which was an ER-2 from Armstrong Flight Research Center (AFRC) in Palmdale California, the TEPC and ARMAS were mounted in their own experiment racks. On commercial aircraft the two instruments were carried in their own cases in the overhead bin on the aircraft. Figures 8 and 9 show the method of carrying the ARMAS and TEPC instruments, respectively, during the flight tests. In addition, Figure 8 shows the ARMAS app running on an iPhone that is paired to the ARMAS instrument. Each instrument carried its own power supply for the flight and was not connected to aircraft systems.

ARMAS FM Calibration to TEPC Utilizing Ground-Based Experimental Results
External to the ARMAS FM's microprocessor, the ARMAS v10.21 algorithms on the ground server transform the Level 0 (L0) millivolt data into L1 D (Si), which is converted to Rad (0.01 Gray) and then to SI units of micro Gray (μGy) using the conversion Equation 3. D Si ð Þ ¼ ch1•14:0 ð Þ× 10 −2 þ ch2•3:6 ð Þ× 10 þ ch3•0:9 ð Þ× 10 4 ; where ch1, ch2, and ch3 are the first three channels of the Teledyne uDOS001. The pseudo-logarithmic output in channel 4 is not used in the data calculation. The L1 measurements of D (Si) are combined with time tag plus location information. The L2 absorbed dose rate, dD (Si)/dt in μGy h −1 , is the time derivative of the L1 TID. Operationally, a 10 data record time step, dt, is used to calculate a rate.

Space Weather
Both the D (Si) and dD (Si)/dt, which are the fundamental measurements by ARMAS, are converted to D (Ti) and dD (Ti)/dt for absorbed dose rate in tissue. This is accomplished by comparing the D (Si) from ARMAS to the D (Ti) of TEPC. A combination of ground-based beamline and flight experiments with both ARMAS and TEPC allowed the calibration then validation between dD (Si)/dt and dD (Ti)/dt. The methodology for representing dose rate in tissue using silicon-based detectors has three steps: (1) conduct the beamline experiments in known, single-species radiation fields and compare the two detectors to find the scaling ratio from dose in silicon to dose in tissue; (2) develop an algorithm to account for weighting the contributions from each species in mixed radiation fields at unique altitudes and magnetic field cutoff rigidities; and (3) validate the D (Si) to D (Ti) conversion by comparing derived ARMAS D (Ti) with measured TEPC D (Ti) using real flight data.

Step 1
Starting with the beamline experiments, we compared the absorbed dose in silicon from ARMAS with the TEPC absorbed dose in tissue. Tables 1-4 show the details for each of the beamline experiments. In the table title each facility, the species measured, and the energies used in the experiment are identified. In each row is presented the ARMAS detector's normalized absorbed dose in silicon in that facility's beamline, the TEPC detector's normalized absorbed dose in tissue in the same beamline, and the scaling ratio required to convert the ARMAS dose in silicon to dose in tissue for that unique source. We use one table for each of the beamline facilities, which have been described above.
A key comparison element is that the only parameter we need to determine the conversion is the ratio between the detectors' normalized observations; they use common units that are unique for different facilities. For example, LANSCE units of measurement were μGy/neutron, LLUMC were μGy/Tic1foil, NSRL were μGy/particle/cm 2 , and LLNL were μGy/h. Since only ratios were needed for scaling ARMAS dose in silicon to TEPC dose in tissue using detectors in the same beamline with common units, this simplified our analysis. The silicon-to-tissue ratio vector for weighting each radiation source is defined in Equation 4 as for 56 Fe ions, alphas, neutrons, protons, gamma rays, electrons, and pions-muons, respectively. C0 = 0.91 • (1/7), where 1/7 normalizes the ratios for seven fractional sources dependent on cutoff rigidity, R c , whose units are GV, and altitude, z, with km used here. The factor of 0.91 is the scaling compensation for 9% unknown error in the calibration of D (Si) to D (Ti) that comes from unmeasured and uncalibrated contributions from known physical processes, including energy from unmodeled electrons and pionsmuons. The elements of the R i ratio vector are composed of the ratios listed in Tables 1-4 converting dose in silicon to dose in tissue. Alpha particles are set to 56 Fe ion values while electrons as well as pions-muons are set to gamma ray photon values since these have not been separately measured but have similar (International Commission on Radiological Protection, 1991) weighting values as described below.

Step 2
Because dose in silicon to dose in tissue ratios for single species radiation fields have now been calculated, the task is to account for the proper mixing of these sources in radiation fields as they occur in the natural atmospheric environment. Using the information for the fractional source contributions based on flight altitude, z, radiation source, i, and a cutoff rigidity, R c , a tensor is formed, GV z,i,Rc , such that a subset can then be iteratively extracted as a vector based on the ith values at an altitude and a unique cutoff rigidity. This vector is called G i and is multiplied by the R i vector at each flight altitude then summed to create a source weighted factor at a flight altitude and cutoff rigidity unique to that flight. This value, called C3 in the ARMAS algorithm, is shown in Equation 5.
The C3 scale factor is used for the entire flight. By using the single flight value at altitude maximum, the operational complexity of calculating C3  introduced during long flights crossing many altitudes and cutoff rigidity levels can be more easily managed as a simplifying approach to calculate flight dose parameter information. Uncertainty at any specific location is, of course, increased. However, the uncertainty is still small since flights typically spend most of their time at cruise (maximum) altitudes and radiation at lower altitudes during take-off and landing is not only much less but occurs for a shorter duration.
The ARMAS D (Si) and dD (Si)/dt are multiplied by the C3 scale factor to get L2 dosimetric parameters of D (Ti) and dD (Ti)/dt that now correspond to the TEPC measured values as shown in Equations 6a and 6b. Across the entire ARMAS database, for all 751 flights at the time of this paper, the median of all the C3 scale factors is 1.42774 ± 0.0822 (1-σ). The variation across all flights had a maximum value in C3 of 1.52702 and a minimum value of 0.96557.
Therefore, with the calculation of D (Ti) and dD (Ti)/dt, the L3 dosimetric parameter of dose equivalent rate, dH/dt in μSv h −1 , can be calculated and is shown in Equations 7a and 7b.
where the Quality factor at the geophysical location of measurement, Q SET , is derived. Q SET uses a fit to the results of Q vs. cutoff rigidity (R c ) by the method in Figure 10 using Equation 8. Q SET has a RMS 1-σ uncertainty of ±12% due to (i) ± 10% uncertainty for modeling R c with the planetary geomagnetic index, Kp; (ii) ±3.6% uncertainty from TEPC-measured, same altitude polar to equatorial estimates (Burda et al., 2013); and (iii) ±5% uncertainty in a polynomial fit to Burda data shown in Figure 10 and Equation 8.
The method using Equation 8 to estimate Q SET was verified by TEPC measurements of Q (Gersey et al., 2012). However, one of the largest uncertainty terms in the ARMAS v10.21 algorithm is for determining the correct Quality factor.
A next L3 dosimetric parameter of ambient dose equivalent rate, dH*(10)/dt in μSv h −1 , is calculated in Equation 9 by multiplying dH/dt with scaling factors to compare favorably with NAIRAS v1 data for commercial aviation altitudes. Ambient Dose Equivalent, H*(10), is a quantity developed for operational field measurements. It reports the average absorbed dose from all radiation at a depth of 10 mm inside a tissue equivalent material such as a human torso phantom. Units for Ambient Dose Equivalent are reported in Sieverts (Sv). NAIRAS v1 calibration was based on the RaD-X balloon mission results .
Here C2 = 1.90 and is the conversion to ambient dose equivalent and C1N = 0.81 is the additional scaling considering dose at a depth of 10-cm in a phantom torso (C. Mertens, private communication).
The final L4 dosimetric parameter of effective dose rate, dE/dt in μSv h −1 , is calculated in Equation 10. Effective Dose, E is derived using a mathematical system that weights the H T (Dose Equivalent received by each separate organ tissue (T) in the human body) by a unique weighting factor (W T ). This weighting factor takes into account the specific sensitivity of each organ to different types of radiation. When the product of these calculations for each organ are summed, the total value is the Effective Dose,

Space Weather
E. Calculating the Effective Dose is especially useful in determining radiation risk for individuals that have received partial body irradiations. For individuals receiving uniform full body irradiations (nonlocalized partial body irradiations), the risk calculated by Effective Dose is the same as that measured in Dose Equivalent. Units for Effective Dose are reported in Sieverts (Sv).
by considering the combined effects of (i) source particle fractional contribution to effective dose at altitude and cutoff rigidity and (ii) the source radiation weighting factor (International Commission on Radiological Protection, 1991). We note that F (z,i) is an array of the fractional contributions by altitude, z, and source species, i, at a specific cutoff rigidity (R c ); W i is an array of ICRP radiation weighting factors by seven sources = [20, 20, 5, 5, 1, 1, 1] where the International Commission on Radiological Protection (1991) values for weighting are heavy ions (e.g., 56 Fe) and alphas = 20, neutrons and protons = 5, and photons, electrons, pions-muons = 1; and C4 = 0.44, which is the weighted scaling to NAIRAS v1 effective dose rate mostly formed by the incoming GCR environment.
Previous calibration discussions for the ARMAS system have been described (Tobiska et al., , 2019 and have been encapsulated in ARMAS algorithms preceding v10. The earlier work  calibrated dD (Ti)/dt and dH/dt dosimetric quantities in the ARMAS system to the RaD-X balloon mission results (Mertens, 2016;Mertens et al., 2016) and dH*(10)/dt as well as dE/dt to the NAIRAS v1 physicsbased radiation model. A summary of that calibration work by Tobiska et al. (2019) provides a discussion of comparative measurements between ARMAS and TEPC between 2011 and 2019.
In this paper we have accomplished more than an updated calibration. We demonstrate the first definitive measurements, in a controlled laboratory environment, of ratios that justify the fit of the ARMAS FM to TEPC for absorbed dose in tissue. In developing this methodology, we have taken the ground-based dosimetry calibration ratios and considered the weighting ratios for atmospheric species and the weighting ratios for species' tissue damage. In the next section, we validate the ARMAS calibration to TEPC measurements for the dD (Ti)/dt and dH/dt parameters in the actual atmosphere modulated by space weather with real flight data. For dH*(10)/dt and dE/dt we continue to use the method of calibrating to NAIRAS v1 . We do not consider shielding by the ARMAS FM housing thickness (100 mil = 0.254 cm aluminum) nor the thickness of the airplane walls, where we use 5.3 g cm −2 aluminum and 3.0 g cm −2 high-density polyethylene (HDPE) (personal communication, March 2010, W. Atwell). Shielding studies with NAIRAS v2 will be done in a separate paper.
The ARMAS v10.21 data processing software made improvements from earlier versions by specifying the relative contribution at altitude for seven sources (heavy ions, alphas, neutrons, protons, photons, electrons, pionsmuons) as used in Equation 10. These sources have a dependence on altitude and cutoff rigidity; examples are shown for cutoff rigidity value in units of GV = 0 (magnetic pole; Figure 11) and GV = 17 (magnetic equator; Figure 12). The fractions of source contributions for a given altitude

10.1029/2020SW002599
Space Weather slice sum to 1.0. In this analysis, we have not included any dependencies in the GCR modulation due to both solar cycle and Forbush decreases. These are important and need to be addressed in future work. The data points between 0 and 13 km were taken from Matthiä et al. (2014) while data points between 14 and 100 km were taken from Norman et al. (2016). Fitted curves to the datapoints are obtained for the effective dose fractional contributions and we use the smooth analytic curves for the ARMAS dose calculations. Separate from those works, alphas and heavies were separated to match the International Commission on Radiological Protection (1991) weighting and, in operations, altitudes greater than 100 km are set to 100 km values. The dose calculations in ARMAS to an altitude of 100 km is important since this marks the top of the atmosphere for suborbital commercial rockets at the present time and SET supplies dose data to those types of organizations in flight support. The Equation 10 based on Figures 11 and 12 allows for a seamless dose exposure environment from the surface of the Earth to the top of the atmosphere.

Step 3
The final task is to validate flight values for ARMAS with TEPC in the actual atmosphere. We have selected seven flights for this comparison where both instruments flew together, mostly in the configuration of Figures 8 and 9 described above. It was noted above that a key scaling factor in this work, C3, converting dose in silicon to dose in tissue was a median value (1.42774 ± 0.0822) across all 751 flights in the ARMAS database. The seven comparison flights were part of this calculation, comprising~1% of the data. The exact C3 values for each flight are listed in Table 5 and are all well within the ±1-σ bounds.
Previously, Tobiska et al. (2019) showed a comparative example of absorbed dose rates for two of the commercial flights used here, i.e., Los Angeles-Denver (20 May 2018 at 11.3 km; Figure 13) and Denver-Los Angeles (23 May 2018 at 11.6 km; Figure 14) during very similar quiet geomagnetic conditions. Flight altitudes and locations come from ADS-B uncorrected pressure altitude flight records that are distributed by FlightAware. These figures have been updated using the new dose rates from the ARMAS v10.21 algorithm. In the top panel of each figure the ARMAS in flight absorbed dose rate, dD (Ti)/dt, is shown vs. time. Error bars of the data are provided and the plot legend provides the numerical value of the RMS uncertainty   Above 20 km on the grid, the notional proton precipitation is shown and color-coded with the cutoff rigidity color bar at the bottom of the plot. The size of the proton representation is proportional to the climatological flux on those field lines. The intent of this proton "ceiling" is to graphically represent the change in cutoff rigidities along the shadowed flight path. Cutoff rigidity modeling is from an updated algorithm provided by Shea and Smart (Shea et al., 1987).

Space Weather
Independent methods of determining Q were used where TEPC provides a measurement while ARMAS makes a calculation based on cutoff rigidities as summarized in Equation 8 and discussed in Tobiska et al. (2016Tobiska et al. ( , 2019. The total ±1σ RMS standard deviation in the ARMAS dose rate dD (Ti)/dt is ±21% and the ±1σ RMS standard deviation for the dose equivalent rate dH/dt is ±24%. The uncertainty in the ambient dose equivalent rate dH*(10)/dt is ±26% and the uncertainty in the effective dose rate dE/dt is ±26%. The terms included in these error calculations are described previously in detail by Tobiska et al. (2016) although we have improved their error estimations in the ARMAS v10.21 algorithm. The uncertainties used in this paper are identified in Table 6.
Typically, minimum cruise altitude values of ARMAS match very well with NAIRAS v1, which is realistic since NAIRAS v1 presents climatology based mostly on the GCR source component. There are very few SEP events in NAIRAS v1, which would be present during GLEs and perhaps subGLEs. These events require further study. Radiation belt particles are not modeled. If there is excess radiation not modeled by NAIRAS v1, as we suspect with bremsstrahlung gamma rays due to precipitated high-energy radiation belt particles, then ARMAS accurately captures that additional dose above the NAIRAS v1 GCR background. Radiation belt particle precipitation to mesospheric altitudes and subsequent gamma ray bremsstrahlung photon production is an area of active research and is not discussed in this paper.

Conclusions
The need for monitoring the radiation environment in aerospace domains relevant to commercial aviation and commercial suborbital travel comes as there is a move toward regulatory compliance for radiation hazard management using guidelines by international organizations such as ICAO. The most important dosimetric parameters for this monitoring include the total absorbed dose (and rate) in tissue as well as the total dose equivalent (and rate). Because the TEPC is a large, weighty instrument not suitable for making real-time measurements, and whereas operational systems need automated, calibrated real-time data feeds, we were motivated in this work to calibrate the ARMAS system to TEPC for the benefit of operational users requiring monitoring compliance. To improve ability for this monitoring compliance, we present, in this study, the silicon-based ARMAS FM cross-calibration with a TEPC that converts dose measured in silicon to dose measured in tissue using four ground-based beamline experiments with species-dependent scaling ratios. Seven aircraft flights were used to validate the derived ARMAS dD (Ti)/dt and dH/dt in comparison with the TEPC measured parameters. In addition, we describe a methodology for normalizing these ratios between different experimental facilities and species. We also provide a methodology for considering mixed radiation fields in the natural atmosphere environment using all the dominant source species that are present. These fractions of species in these mixed radiation fields are dynamic and vary with altitude as well as magnetic field cutoff rigidity, which is continually perturbed by the solar wind.
In the first of three steps, i.e., the development of scaling ratios, both ARMAS and TEPC detectors were exposed to high energy and relativistic neutrons at the Los Alamos Neutron Science Center (LANSCE), high energy protons at Loma Linda University Medical Center (LLUMC), high energy 56 Fe particles at the NASA Space Radiation Laboratory (NSRL) Brookhaven National Laboratory, and a 60 Co gamma ray source at Lawrence Livermore National Laboratory. These four radiation sources were chosen to represent the main primary and secondary components of the mixed radiation field that occurs naturally from 8 to 100 km altitude.
In the second of three steps, results from the ground-based exposures were used to develop a normalization methodology using a species-dependent scaling ratio that includes consideration of species' fractional contributions at altitude and modulated by cutoff rigidity. This allowed the determination of D (Ti) from the

Space Weather
ARMAS FM D (Si) measurement in realistic atmospheric conditions. Combined with a previously verified quadratic function that creates a Quality factor, Q SET , based on cutoff rigidity, the dose equivalent rate, dH/dt, is then calculated to provide regulatory compliance relevant dose exposure rates on aircraft and suborbital vehicles.
In the third of three steps, the ARMAS FM and the TEPC detectors were flown together on commercial and research aircraft to validate the derived ARMAS dH/dt results with the TEPC measured dH/dt. The ARMAS v10.21 algorithm was applied to ARMAS FM flight measurements for seven continental U.S. flights where TEPC was also simultaneously flown. The ARMAS results compared very favorably with the "gold standard" TEPC flight measurements; the TEPC dH/dt results were within ±1σ RMS standard deviation of the ARMAS derived dH/dt values.
The ARMAS system provides real-time dosimetric parameters via Iridium satellite link in research aircraft and high-altitude balloons as well as via Bluetooth pairing to the ARMAS iPhone and iPad app in support of radiation situational awareness and monitoring. This system provides, for the first time, a solid foundation for using calibrated, automated silicon-based detectors to monitor compliance of dosimetric parameters with guidelines that are considered important by ICAO.