Sb. Curtis et al., RISK CROSS-SECTIONS AND THEIR APPLICATION TO RISK-ESTIMATION IN THE GALACTIC COSMIC-RAY ENVIRONMENT, Radiation research, 141(1), 1995, pp. 57-65
Radiation risk cross sections (i.e, risks per particle fluence) are di
scussed in the context of estimating the risk of radiation-induced can
cer on long-term space flights from the galactic cosmic radiation outs
ide the confines of the earth's magnetic field. Such quantities are us
eful for handling effects not seen after low-LET radiation. Since appr
opriate cross-section functions for cancer induction for each particle
species are not yet available, the conventional quality factor is use
d as an approximation to obtain numerical results for risks of excess
cancer mortality. Risks are obtained for seven of the most radiosensit
ive organs as determined by the ICRP [stomach, colon, lung, bone marro
w (BFO), bladder, esophagus and breast], beneath 10 g/cm(2) aluminum s
hielding at solar minimum. Spectra are obtained for excess relative ri
sk for each cancer per LET interval by calculating the average fluence
-LET spectrum for the organ and converting to risk by multiplying by a
factor proportional to R(gamma) L Q(L) before integrating over L, the
unrestricted LET. Here R(gamma) is the risk coefficient for low-LET r
adiation (excess relative mortality per Sv) for the particular organ i
n question. The total risks of excess cancer mortality obtained are 1.
3 and 1.1% to female and male crew, respectively, for a 1-year exposur
e at solar minimum. Uncertainties in these values are estimated to ran
ge between factors of 4 and 15 and are dominated by the biological unc
ertainties in the risk coefficients for low-LET radiation and in the L
ET (or energy) dependence of the risk cross sections (as approximated
by the quality factor). The direct substitution of appropriate risk cr
oss sections will eventually circumvent entirely the need to calculate
, measure or use absorbed dose, equivalent dose and quality factor for
such a high-energy charged-particle environment.