Chapter 29 — Radiation in the Environment: Natural Background, Accidents, and Protection

"Radiation is one of the important facts of life. Our bodies are radioactive. The ground we stand on is radioactive. The food we eat, the air we breathe, and the water we drink are all slightly radioactive." — Merril Eisenbud, Environmental Radioactivity (1963)

Chapter Overview

Every second of every day, roughly 4,400 atoms of ${}^{40}\text{K}$ decay inside your body. Cosmic ray muons pass through you at a rate of about one per square centimeter per minute. The air in your basement likely contains trace amounts of ${}^{222}\text{Rn}$, an alpha-emitting noble gas that is the single largest source of radiation exposure for most people on Earth. You are, and always have been, a radioactive being living in a radioactive world.

This chapter brings together the nuclear physics developed in Chapters 12 (radioactivity fundamentals) and 16 (radiation interactions with matter) to address three questions that connect nuclear science to public health, policy, and daily life:

1. Where does environmental radiation come from? We will inventory the natural and man-made sources, quantify each contribution, and discover that natural sources and medical procedures — not nuclear power plants or weapons testing — dominate the radiation budget for most people.

2. What does radiation do to living tissue? We will distinguish deterministic effects (which appear above a dose threshold and whose severity increases with dose) from stochastic effects (principally cancer, whose probability — but not severity — increases with dose), and confront the deep scientific uncertainty about effects at low doses.

3. How do we protect ourselves? We will develop the principles of radiation protection — justification, optimization, and dose limitation — and examine the technologies used to measure and monitor radiation exposure.

Throughout, we will be rigorously quantitative but honest about uncertainty. The physics of radiation — its production, transport, and energy deposition — is well understood. The biology of low-dose radiation is not. The chapter title says "protection," and the nuclear physicist's contribution to protection begins with getting the physics right.

Spaced Review — Chapter 12 Connection: Recall from Chapter 12 that the activity of a sample $A = \lambda N = (\ln 2 / t_{1/2}) N$ measures the number of decays per second (in becquerels). Secular equilibrium ensures that in undisturbed natural decay chains (e.g., ${}^{238}\text{U} \to \cdots \to {}^{222}\text{Rn} \to \cdots \to {}^{206}\text{Pb}$), every daughter has the same activity as the long-lived parent. This is why radon activity in soil is directly determined by the uranium content of the underlying rock.

Spaced Review — Chapter 16 Connection: Chapter 16 introduced the radiation dose units we will use throughout: the gray (Gy, absorbed energy per unit mass, 1 Gy = 1 J/kg) and the sievert (Sv, the gray weighted by the radiation quality factor to reflect biological effectiveness). Chapter 16 also derived the Bethe-Bloch formula for charged-particle energy loss and the interaction coefficients for photons — the physics that determines how radiation deposits energy in tissue.

🏃 Fast Track: If you are comfortable with radioactivity and dose units, skim Sections 29.1–29.2 and begin at Section 29.4 (biological effects). The essential new physics is in Sections 29.4–29.6 (biological effects, LNT debate, and protection principles).

🔬 Deep Dive: The LNT debate (Section 29.5) is one of the most consequential unresolved questions in radiation science. Engaging with the primary literature cited there will prepare you for graduate-level health physics and radiation biology.

29.1 Natural Radioactivity: The Baseline We Inherited

The Earth formed 4.54 billion years ago from material that was already radioactive. The heaviest elements — uranium, thorium, and their daughters — were synthesized in the r-process of neutron star mergers and supernovae (Chapter 23) and incorporated into the solar nebula. These primordial radionuclides, along with isotopes continuously produced by cosmic ray interactions in the atmosphere, establish the natural background radiation that has bathed all life on Earth since its origin.

29.1.1 Terrestrial Sources: Uranium, Thorium, and Potassium

Three long-lived radionuclides dominate the terrestrial contribution to natural background radiation:

Uranium-238 heads a decay chain of 14 steps (8 alpha decays, 6 beta decays) terminating at stable ${}^{206}\text{Pb}$. Each step produces radiation — alpha particles, beta particles, gamma rays, or all three. In secular equilibrium, every member of the chain has the same activity as the parent. In a kilogram of typical soil containing 25 Bq of ${}^{238}\text{U}$, the total activity from the entire chain is $14 \times 25 = 350$ Bq. (In practice, equilibrium is partially disrupted by the escape of gaseous ${}^{222}\text{Rn}$, as we discuss in Section 29.3.)

Thorium-232 heads a 10-step chain terminating at ${}^{208}\text{Pb}$. Its longer half-life means more atoms are needed for the same activity, but higher crustal abundance compensates.

Potassium-40 is a single-step decay: 89.3% of the time it undergoes $\beta^-$ decay to ${}^{40}\text{Ca}$ (endpoint energy 1.311 MeV), and 10.7% undergoes electron capture to ${}^{40}\text{Ar}$ (followed by a characteristic 1.461 MeV gamma ray). Despite its relatively low isotopic abundance — only 0.0117% of natural potassium is ${}^{40}\text{K}$ — potassium is so abundant in rocks and soil that ${}^{40}\text{K}$ is the dominant single contributor to external terrestrial gamma-ray exposure.

💡 Numerical check: Potassium constitutes about 2.1% of the Earth's crust by mass. In 1 kg of soil, there is about 21 g of K, of which $0.0117\% = 2.46 \times 10^{-3}$ g is ${}^{40}\text{K}$. Converting:

$$N_{40} = \frac{2.46 \times 10^{-3}\,\text{g}}{40\,\text{g/mol}} \times 6.022 \times 10^{23}\,\text{mol}^{-1} = 3.7 \times 10^{19}\,\text{atoms}$$

$$A = \frac{\ln 2}{1.248 \times 10^9 \times 3.156 \times 10^7\,\text{s}} \times 3.7 \times 10^{19} = 650\,\text{Bq}$$

This is consistent with the measured range of 370–700 Bq/kg for typical soils, validating our simple estimate.

The external dose rate from terrestrial gamma radiation depends strongly on local geology. Granitic regions (high U/Th content) deliver higher doses than sedimentary basins. Worldwide averages, compiled by UNSCEAR (the United Nations Scientific Committee on the Effects of Atomic Radiation), are:

$$\dot{D}_{\text{terrestrial, external}} \approx 0.48\,\text{mSv/yr}$$

with a range of roughly 0.3–1.0 mSv/yr for most inhabited regions. Exceptional hotspots exist:

These natural high-background areas have been extensively studied for health effects. We will return to them in Section 29.5 when we discuss the LNT debate.

29.1.2 Internal Sources: You Are Radioactive

Radionuclides enter the body through food, water, and inhalation and establish an internal radiation dose. The dominant internal emitter is ${}^{40}\text{K}$.

Potassium-40 in the body: The human body maintains approximately 140 g of potassium through homeostatic regulation — regardless of dietary intake, the body adjusts to maintain a concentration of about 2 g K per kg of body mass. For a 70 kg person:

$$N_{40} = \frac{0.0117\% \times 140\,\text{g}}{40\,\text{g/mol}} \times 6.022 \times 10^{23} = 2.47 \times 10^{20}\,\text{atoms}$$

$$A = \frac{\ln 2}{1.248 \times 10^9 \times 3.156 \times 10^7\,\text{s}} \times 2.47 \times 10^{20} = \boxed{4{,}400\,\text{Bq}}$$

You emit roughly 4,400 beta particles and gamma rays per second from potassium alone. The resulting internal dose is approximately 0.17 mSv/yr, dominated by the beta component (the gammas mostly escape the body).

📊 Reality check: This means every human being is a radioactive source with an activity of about 4,400 Bq. You cannot reduce this — potassium is an essential electrolyte, and the body's homeostatic mechanisms maintain its concentration regardless of intake. If you eat a banana (which contains about 15 Bq of ${}^{40}\text{K}$), the excess potassium is excreted within hours. The "banana equivalent dose" sometimes used in popular science is therefore misleading: eating a banana does not increase your body's total ${}^{40}\text{K}$ activity.

Other internal radionuclides contribute smaller doses:

The ${}^{210}\text{Po}$ contribution merits attention. Polonium-210 is an alpha emitter ($E_\alpha = 5.30$ MeV, $t_{1/2} = 138$ d) at the bottom of the ${}^{238}\text{U}$ chain. It accumulates in the body through inhalation of ${}^{222}\text{Rn}$ daughters and through diet (especially seafood and tobacco — the trichomes of tobacco leaves absorb ${}^{210}\text{Pb}$ from the air, which decays in situ to ${}^{210}\text{Po}$). Smokers receive a localized lung dose from ${}^{210}\text{Po}$ that is several times higher than for non-smokers.

29.1.3 Cosmogenic Radionuclides

Cosmic rays — primarily high-energy protons and alpha particles from galactic and solar sources — continuously bombard the upper atmosphere. Nuclear spallation reactions produce a suite of radioactive isotopes:

Carbon-14 is the most important cosmogenic radionuclide. It is produced when cosmic-ray-generated neutrons capture on ${}^{14}\text{N}$ — the dominant constituent of the atmosphere. The ${}^{14}\text{C}$ oxidizes to ${}^{14}\text{CO}_2$, enters the carbon cycle, and is incorporated into all living organisms. In secular equilibrium, the specific activity of carbon in living tissue is approximately 226 Bq/kg C (or about 0.264 Bq per gram of total body mass). After death, ${}^{14}\text{C}$ decays without replenishment — the basis of radiocarbon dating (Chapter 12).

🔗 Chapter 12 Connection: The radiocarbon dating equation $t = 8{,}267\,\text{yr} \times \ln(A_0/A)$ was derived in Section 12.8. Here we see the nuclear physics behind $A_0$: it is set by the cosmic ray production rate, the atmospheric ${}^{14}\text{C}/{}^{12}\text{C}$ ratio ($\sim 1.2 \times 10^{-12}$), and the mixing time of the carbon cycle.

29.2 Cosmic Rays: Radiation from the Sky

29.2.1 The Cosmic Ray Spectrum

Primary cosmic rays striking the top of the atmosphere are approximately 87% protons, 12% alpha particles, and 1% heavier nuclei, with energies spanning from $\sim 10^8$ eV to beyond $10^{20}$ eV. When these primaries collide with atmospheric nitrogen and oxygen nuclei, they produce extensive air showers of secondary particles: pions, kaons, muons, electrons, photons, and neutrons.

At sea level, the dominant component of the secondary cosmic ray flux is muons (produced by pion decay high in the atmosphere). The muon flux at sea level is approximately:

$$\Phi_\mu \approx 1\,\text{cm}^{-2}\,\text{min}^{-1} \approx 170\,\text{m}^{-2}\,\text{s}^{-1}$$

Each muon carries a typical energy of $\sim 3$ GeV and loses energy primarily through ionization (Bethe-Bloch, Chapter 16). The resulting dose rate at sea level from all cosmic ray secondaries is approximately:

$$\dot{D}_{\text{cosmic, sea level}} \approx 0.34\,\text{mSv/yr}$$

29.2.2 Altitude Dependence

The atmosphere provides approximately 1,030 g/cm$^2$ of shielding (about 10 mean free paths for hadronic interactions). The cosmic ray dose rate increases roughly exponentially with altitude as this shielding is removed:

$$\dot{D}(h) \approx \dot{D}_0 \, e^{h/h_0}$$

where $h_0 \approx 1{,}500$ m is the scale height and $\dot{D}_0$ is the sea-level rate. Representative values:

⚠️ Practical consequence: At jet cruising altitude of 10,700 m (35,000 ft), the cosmic ray dose rate is approximately 5 $\mu$Sv/hr — roughly 100 times the sea-level rate. Airline crew who fly long-haul routes accumulate 2–5 mSv/yr from cosmic radiation alone, making them one of the most radiation-exposed occupational groups. This is comparable to the dose limit recommended for the general public (1 mSv/yr above background) and motivates dedicated dosimetry programs for aircrew in the EU and other jurisdictions.

29.2.3 The South Atlantic Anomaly and Solar Modulation

The Earth's magnetic field deflects low-energy cosmic rays, providing geomagnetic shielding that is strongest at the equator and weakest near the magnetic poles. The South Atlantic Anomaly (SAA) — a region where the inner Van Allen radiation belt dips closest to Earth's surface due to the offset between the geomagnetic and geographic axes — produces enhanced radiation exposure for aircraft and spacecraft passing through this zone over the South Atlantic.

The cosmic ray flux also varies with the 11-year solar cycle: during solar maximum, the enhanced solar wind deflects galactic cosmic rays, reducing the sea-level neutron flux by 15–25% compared to solar minimum. This solar modulation is clearly visible in neutron monitor data and must be accounted for in precise dosimetry.

29.3 Radon: The Invisible Threat in Your Basement

29.3.1 Origin and Nuclear Physics

Radon-222 is a noble gas produced by the alpha decay of ${}^{226}\text{Ra}$ (radium-226), which is itself part of the ${}^{238}\text{U}$ decay chain:

$${}^{238}\text{U} \xrightarrow{4.47\,\text{Gyr}} {}^{234}\text{Th} \xrightarrow{\cdots} {}^{226}\text{Ra} \xrightarrow{1600\,\text{yr}} {}^{222}\text{Rn} \xrightarrow{3.82\,\text{d}} {}^{218}\text{Po} \xrightarrow{\cdots} {}^{206}\text{Pb}$$

The key nuclear physics:

• ${}^{222}\text{Rn}$ has a half-life of 3.824 days — long enough to diffuse through soil and rock into buildings, but short enough that it does not accumulate in the outdoor atmosphere (where it is diluted to negligible concentrations).
• ${}^{222}\text{Rn}$ itself is a noble gas and chemically inert. It is inhaled and exhaled without significant deposition in the lungs.
• The daughters of ${}^{222}\text{Rn}$ are the real hazard. When ${}^{222}\text{Rn}$ decays in the air, its daughter ${}^{218}\text{Po}$ is a charged metal atom that rapidly attaches to aerosol particles or surfaces. These daughters — ${}^{218}\text{Po}$ ($\alpha$, 3.10 min), ${}^{214}\text{Pb}$ ($\beta^-$, 26.8 min), ${}^{214}\text{Bi}$ ($\beta^-$, 19.9 min), ${}^{214}\text{Po}$ ($\alpha$, 164 $\mu$s) — deposit in the bronchial epithelium when inhaled, delivering intense alpha radiation to the radiosensitive basal cells of the lung.

🔗 Chapter 12 Connection: In the ${}^{222}\text{Rn}$ sub-chain, the first daughter ${}^{218}\text{Po}$ ($t_{1/2} = 3.10$ min) is in transient equilibrium with radon. But ${}^{214}\text{Po}$ ($t_{1/2} = 164\,\mu$s) is in secular equilibrium with its parent ${}^{214}\text{Bi}$ ($t_{1/2} = 19.9$ min). The Bateman equations (Section 12.5) govern the exact activity ratios at every point in this chain.

29.3.2 Radon as the Dominant Natural Radiation Source

According to NCRP Report No. 160 (2009) — the definitive source for US radiation exposure data — the average annual effective dose to the US population from all sources is approximately:

$$D_{\text{total}} \approx 6.2\,\text{mSv/yr}$$

The breakdown:

Radon inhalation alone accounts for 37% of the total — the single largest contributor. When we discuss "natural background," radon dominates; yet its concentration varies enormously from building to building, making individual exposure highly variable.

29.3.3 Radon Concentrations and Building Physics

Radon enters buildings primarily through cracks in foundations, gaps around pipes, and unsealed sump pits. The indoor concentration depends on:

1. Soil uranium/radium content — the ultimate source
2. Soil permeability — governs gas transport
3. Building construction — foundation type, sealing, ventilation
4. Stack effect — warm air rising inside creates slight negative pressure at the basement level, drawing soil gas in

Indoor radon concentrations are measured in becquerels per cubic meter (Bq/m$^3$) or, in the US, in picocuries per liter (pCi/L), where $1\,\text{pCi/L} = 37\,\text{Bq/m}^3$.

The EPA's action level for residential radon is 4 pCi/L (148 Bq/m$^3$) — at or above this concentration, the EPA recommends mitigation. The WHO recommends a lower reference level of 100 Bq/m$^3$ (2.7 pCi/L).

Representative radon concentrations:

The Watras case is famous: in December 1984, Stanley Watras, an engineer at the Limerick nuclear power plant in Pennsylvania, repeatedly set off the plant's radiation monitors — and the plant had not yet loaded fuel. The source was eventually traced to his home, which had a basement radon concentration nearly 700 times the EPA action level. The discovery triggered a national survey that revealed radon as a widespread indoor air quality problem.

29.3.4 Radon and Lung Cancer: The Epidemiological Evidence

The causal link between radon and lung cancer is established beyond reasonable doubt, primarily from studies of underground miners:

1. Miner cohort studies: The most important data come from 11 pooled cohort studies covering over 60,000 underground miners (uranium, tin, iron, fluorspar) followed for decades. These studies demonstrate a clear, statistically significant, dose-response relationship between cumulative radon progeny exposure (measured in Working Level Months, WLM) and excess relative risk of lung cancer.

2. Residential studies: Pooled analyses of case-control studies in Europe (Darby et al., 2005, BMJ) and North America (Krewski et al., 2005, J. Rad. Prot.) show a statistically significant 8–16% increase in lung cancer risk per 100 Bq/m$^3$ of long-term average residential radon concentration.

3. Radon-smoking synergy: Radon and smoking interact synergistically. A smoker exposed to radon has a much higher lung cancer risk than predicted by adding the two risks independently. The BEIR VI report (1999) estimated that radon causes approximately 21,000 lung cancer deaths per year in the US — about 2,900 among never-smokers and 18,100 among current and former smokers.

📊 Putting it in context: Radon is the second-leading cause of lung cancer after smoking, and the leading cause of lung cancer among non-smokers. The EPA estimates that radon causes more deaths per year than drunk driving (~10,000) and roughly as many as breast cancer (~40,000) — yet public awareness remains remarkably low compared to these other risks.

29.3.5 Measurement and Mitigation

Measurement of indoor radon is straightforward and inexpensive:

• Short-term tests (2–7 days): Activated charcoal canisters or alpha-track detectors placed in the lowest livable level. Cost: \$10–\$30.
• Long-term tests (90+ days): Alpha-track detectors that integrate over seasonal variations. More representative of annual average exposure.
• Continuous monitors: Electronic devices that measure radon concentration in real time. Used for diagnostic purposes and after mitigation.

Mitigation is effective and affordable. The standard technique for homes with basements or slab-on-grade foundations is sub-slab depressurization (SSD):

1. A PVC pipe is inserted through the basement slab into the gravel layer beneath.
2. A fan draws soil gas from below the slab and exhausts it above the roofline.
3. The slight negative pressure beneath the slab prevents radon from entering the building.

SSD systems typically reduce indoor radon by 80–99% and cost \$800–\$2,500 for installation, with minimal operating cost (the fan uses about 70–90 W continuously). This is one of the highest-impact, lowest-cost public health interventions available.

29.4 Man-Made Sources of Radiation Exposure

29.4.1 Medical Exposure: The Dominant and Growing Contribution

The most striking feature of modern radiation exposure statistics is the dramatic growth of medical radiation. In 1980, medical radiation contributed about 15% of the average American's dose; by 2006, it contributed nearly 50%. The driver: computed tomography (CT).

The number of CT scans performed in the US grew from approximately 3 million in 1980 to over 80 million in 2020. CT scanning delivers substantially higher doses than conventional radiography:

A single CT abdomen delivers roughly the same effective dose as 500 chest X-rays. This has prompted significant attention to dose optimization in diagnostic radiology — the ALARA principle (Section 29.6) applies to medicine as well as to the nuclear industry.

📊 Key distinction: Medical radiation is almost entirely voluntary and individually beneficial — the diagnostic information from a medically indicated CT scan almost always outweighs the radiation risk. The concern is not with any single justified scan, but with aggregate overuse: the ordering of scans where the clinical benefit is marginal, repeat scanning, and failure to use dose-reduction techniques.

Nuclear medicine contributes a further 0.77 mSv/yr to the US average, primarily from PET scans using ${}^{18}\text{F}$-FDG (Chapter 27) and diagnostic procedures using ${}^{99m}\text{Tc}$.

29.4.2 Nuclear Weapons Testing Fallout

Atmospheric nuclear weapons testing, conducted primarily between 1945 and 1963 (when the Partial Test Ban Treaty drove testing underground), injected substantial quantities of fission products into the global atmosphere. The most important fallout nuclides:

The global ${}^{14}\text{C}$ spike is the most long-lasting signature: atmospheric testing approximately doubled the ${}^{14}\text{C}/{}^{12}\text{C}$ ratio by 1963. This "bomb pulse" is now being incorporated into the biosphere and gradually declining as the excess ${}^{14}\text{C}$ equilibrates with the oceanic and terrestrial carbon reservoirs. The bomb pulse has found unexpected scientific applications: forensic scientists use it to determine the age of biological samples (wines, ivory, human tissues), and it has been used to study the turnover rate of human cells.

Local contamination was far more severe. Fallout from individual tests contaminated downwind areas with ${}^{131}\text{I}$ (causing thyroid doses, especially in children via the grass $\to$ cow $\to$ milk $\to$ child pathway), ${}^{137}\text{Cs}$ (causing external and internal doses through contaminated soil and food), and ${}^{90}\text{Sr}$ (which substitutes for calcium in bone, delivering chronic beta radiation to the bone marrow).

The current (2020s) global dose commitment from residual weapons fallout is approximately 0.005 mSv/yr — negligible compared to natural background. But the local consequences for downwind populations (e.g., Marshall Islands, Semipalatinsk, Nevada Test Site) were severe and remain a matter of ongoing medical follow-up and political reckoning.

29.4.3 Nuclear Power: Normal Operations and Accidents

Normal operations of nuclear power plants contribute negligibly to public radiation exposure. The annual dose to the most-exposed member of the public from a typical PWR is less than 0.01 mSv — well below the 1 mSv/yr public dose limit and roughly equivalent to a few hours of natural background.

🔗 Chapter 26 Connection: The reactor physics of containment, multiple barriers (fuel cladding $\to$ pressure vessel $\to$ containment building), and effluent treatment ensures that routine releases are tiny. The relevant nuclear physics is the fission product inventory discussed in Chapter 26 — a 1 GW$_e$ reactor contains roughly $10^{20}$ Bq of fission products, but multiple engineered barriers prevent their release.

Accidents, however, can produce locally significant contamination. The three major reactor accidents illustrate the spectrum:

Three Mile Island (1979): Partial core meltdown of TMI-2 (PWR) released approximately $4 \times 10^{17}$ Bq of noble gases (${}^{133}\text{Xe}$, ${}^{85}\text{Kr}$) but very little particulate radioactivity. The estimated average dose to the 2 million people within 50 miles was 0.015 mSv — less than one day of natural background. No attributable health effects have been detected.

Chernobyl (1986): The most severe reactor accident in history. The RBMK-1000 design lacked a Western-style containment building; the reactivity excursion and subsequent graphite fire released approximately $1.2 \times 10^{19}$ Bq of radioactive material (excluding noble gases), including large fractions of the core's ${}^{131}\text{I}$ and ${}^{137}\text{Cs}$ inventory. Health consequences included: - 134 cases of acute radiation syndrome (ARS) among plant workers and first responders, of whom 28 died within months - A documented increase in thyroid cancer among those exposed as children (>6,000 cases by 2005), primarily due to ${}^{131}\text{I}$ via the milk pathway — most cases were treatable - The excess cancer risk to the broader population is estimated at a few thousand over a lifetime but is not detectable above the statistical noise of the 25% baseline cancer rate

Fukushima (2011): Three reactor meltdowns following the magnitude 9.0 earthquake and tsunami. Total release approximately $1.5 \times 10^{17}$ Bq of ${}^{131}\text{I}$ and $4 \times 10^{16}$ Bq of ${}^{137}\text{Cs}$ — about one-tenth to one-sixth of the Chernobyl release. No acute radiation deaths. The WHO and UNSCEAR concluded that the radiation doses to the public were generally low and that no discernible increase in cancer rates is expected.

⚠️ A perspective calculation: The approximately 19,500 deaths from the earthquake and tsunami at Fukushima far exceeded any radiation health consequence. However, the evacuation itself — involving over 150,000 people, many elderly — contributed to an estimated 1,000–2,000 excess deaths from displacement stress, interrupted medical care, and suicide. The decision of when to evacuate and when to shelter-in-place is a complex optimization that depends critically on accurate dose projections.

29.4.4 Consumer Products and Other Sources

Smaller man-made contributions include:

• Tobacco: ${}^{210}\text{Po}$ in cigarettes delivers localized alpha doses to the lung. For a pack-a-day smoker, the annual lung dose from ${}^{210}\text{Po}$ is estimated at 80–160 mSv to the bronchial epithelium — far exceeding any other single source.
• Building materials: Concrete, granite, and brick contain trace U/Th/K. Granite countertops are sometimes cited as a radon source, but the contribution is negligible compared to soil.
• Air travel: As noted in Section 29.2.2, each hour at cruising altitude delivers ~5 $\mu$Sv. A transatlantic round trip (New York to London) delivers ~0.06 mSv.
• Coal combustion: Coal contains trace uranium and thorium, and coal ash concentrates these radionuclides. The radiation exposure from living near a coal power plant actually exceeds that from living near a nuclear plant under normal operations.

29.5 Biological Effects of Ionizing Radiation

The nuclear physics of radiation — its production, interactions, and energy deposition — is well understood (Chapters 12–16). The biological consequences are far more complex and, at low doses, genuinely uncertain. This section develops the framework for understanding what radiation does to living tissue.

29.5.1 The Physical and Chemical Stages

When ionizing radiation passes through tissue, the sequence of events spans many orders of magnitude in time:

1. Physical stage ($< 10^{-15}$ s): Energy is deposited through ionization and excitation of water molecules and biomolecules. Alpha particles, protons, and heavy ions produce dense ionization tracks (high linear energy transfer, LET); electrons and gamma rays produce sparse tracks (low LET). Chapter 16 treated this stage in detail.

2. Chemical stage ($10^{-15}$ to $10^{-3}$ s): Ionized water molecules produce free radicals — primarily hydroxyl radicals (OH$\cdot$) — that are highly reactive. Approximately 60–70% of radiation damage to DNA is caused indirectly through free radical attack, rather than by direct ionization of the DNA molecule itself.

3. Biological stage (seconds to decades): DNA damage is either repaired or misrepaired. Base excision repair, nucleotide excision repair, and homologous recombination handle most lesions. The critical damage type is the double-strand break (DSB) — a break in both strands of the DNA helix within a few base pairs. DSBs are difficult to repair and their misrepair can lead to chromosomal aberrations, mutations, or cell death.

💡 Scale of damage: A whole-body dose of 1 Gy of gamma radiation produces approximately 40 DSBs per cell. The cell's repair machinery handles this routinely — endogenous metabolic processes produce roughly 10–50 DSBs per cell per day from oxidative stress alone. The difference with radiation is that the damage is concentrated in time (especially for acute exposure) and, for high-LET radiation, concentrated in space (clustered damage along the particle track).

29.5.2 Deterministic Effects: Above the Threshold

Deterministic effects (also called tissue reactions in modern ICRP terminology) occur when radiation kills enough cells in a tissue to impair function. They are characterized by:

1. A dose threshold below which the effect does not occur (because the tissue's reserve capacity and repair mechanisms compensate for the lost cells)
2. Severity increases with dose above the threshold
3. A short latency (hours to weeks)

The major deterministic effects and their approximate thresholds (for acute, whole-body exposure to low-LET radiation):

29.5.3 Acute Radiation Syndrome (ARS)

Acute radiation syndrome develops after a whole-body dose exceeding approximately 1 Sv delivered over a short period. ARS progresses through characteristic phases that depend on dose:

Hematopoietic syndrome (1–8 Sv): The bone marrow, which contains rapidly dividing stem cells, is the most radiosensitive tissue. Lymphocyte counts drop within hours (the earliest and most sensitive biomarker). Neutrophils and platelets decline over 2–4 weeks as the irradiated stem cells fail to replace the normal turnover. Patients become susceptible to infection (neutropenia) and hemorrhage (thrombocytopenia). With supportive care (antibiotics, transfusions, growth factors), survival is possible up to ~8 Sv.

Gastrointestinal syndrome (8–30 Sv): The epithelial lining of the small intestine, which renews every 3–5 days, is destroyed. The resulting loss of the intestinal barrier leads to fluid loss, electrolyte imbalance, and sepsis as gut bacteria enter the bloodstream. This syndrome is nearly always fatal, typically within 1–2 weeks.

Neurovascular syndrome (>30 Sv): At extremely high doses, direct damage to the central nervous system produces rapid onset of confusion, ataxia, convulsions, and cardiovascular collapse. Death occurs within hours to days. This was the fate of some of the Chernobyl first responders and of criticality accident victims such as Louis Slotin (Los Alamos, 1946) and Hisashi Ouchi (Tokaimura, 1999).

⚠️ A sobering reality: The dose rate matters enormously. The same total dose of 3 Sv delivered over a few minutes produces severe ARS, but the same 3 Sv spread over 30 years (as occupational exposure) produces no deterministic effects at all — the body's repair mechanisms can keep pace when the damage rate is low. This is the fundamental reason why fractionated radiotherapy works: tumor cells repair less efficiently than normal tissue, so spreading the dose over many fractions exploits the differential repair capacity.

29.5.4 Stochastic Effects: The Cancer Question

Stochastic effects — principally cancer — differ fundamentally from deterministic effects:

1. No certain threshold: Even a single radiation-induced DSB, if misrepaired, can in principle initiate a chain of events leading to cancer (though the probability from a single DSB is extremely small)
2. Probability increases with dose, but severity does not — a radiation-induced cancer is clinically indistinguishable from a spontaneously occurring cancer
3. Long latency: Years to decades between exposure and cancer appearance (2–5 years for leukemia, 10–40 years for solid tumors)

The primary evidence for radiation carcinogenesis comes from the Life Span Study (LSS) of the Japanese atomic bomb survivors — the most important epidemiological study in radiation biology. This study has followed approximately 120,000 survivors from 1950 to the present, with individualized dose estimates. Key findings:

• A statistically significant excess of cancer is observed at doses above approximately 100–200 mSv
• The excess relative risk (ERR) for solid cancers is approximately 0.47 per Sv (i.e., a 1 Sv dose increases cancer risk by about 47%)
• For leukemia, the ERR is higher (~1.5 per Sv) and the dose-response is better fit by a linear-quadratic model
• The absolute excess risk varies by cancer type, sex, and age at exposure (those exposed as children have higher lifetime risk)

Quantitative risk estimation: The ICRP (Publication 103, 2007) estimates the nominal risk coefficient for radiation-induced fatal cancer in the whole population as:

$$r \approx 5\% \,\text{per Sv} = 5 \times 10^{-2}\,\text{Sv}^{-1}$$

This means a dose of 1 Sv is estimated to increase the lifetime cancer mortality risk by approximately 5 percentage points, on a baseline of roughly 25% (since about one in four people dies of cancer regardless of radiation exposure). For non-fatal cancers and hereditary effects, the total detriment coefficient is approximately 5.7% per Sv.

💡 Applying the risk coefficient: A chest CT (7 mSv) would carry an estimated excess cancer risk of $7 \times 10^{-3}\,\text{Sv} \times 5 \times 10^{-2}\,\text{Sv}^{-1} = 3.5 \times 10^{-4}$, or about 1 in 2,900. This is a useful framework for clinical risk-benefit analysis, but it relies on the linear no-threshold model, whose validity at this dose level is precisely the question we address next.

29.5 The LNT Debate: What We Know and Don't Know

29.5.1 The Linear No-Threshold Model

The linear no-threshold (LNT) model asserts that:

$$R(D) = R_0 + \alpha D$$

where $R(D)$ is the cancer risk at dose $D$, $R_0$ is the baseline cancer risk, and $\alpha$ is a constant risk coefficient. There is no threshold dose below which radiation is safe, and risk is strictly proportional to dose all the way down to zero.

The LNT model was adopted by the major radiation protection bodies (ICRP, NCRP, BEIR) as the basis for radiation protection standards. Its logical consequence: any dose, no matter how small, carries some risk, and therefore all unnecessary radiation exposure should be minimized.

29.5.2 Evidence Supporting LNT

1. Linear dose-response in the LSS: The dose-response for solid cancers in the atomic bomb survivors is consistent with linearity down to about 100–200 mSv. The BEIR VII committee (2006) concluded that "the linear no-threshold model provides the best fit to the LSS data."

2. Mechanistic plausibility: A single radiation-induced DSB can, in principle, initiate carcinogenesis. There is no known mechanism that would make a cell completely safe from misrepair at any nonzero dose.

3. Conservation: The LNT model does not underestimate risk. If the true dose-response curves downward at low doses (as some evidence suggests), then LNT provides a conservative upper bound on risk — a desirable property for a radiation protection framework.

4. Simplicity and additivity: LNT allows straightforward addition of doses from multiple sources and time periods, enabling practical radiation protection management.

29.5.3 Evidence Against LNT (or at Least Questioning It)

1. The hormesis hypothesis: Some studies suggest that low doses of radiation may actually be protective — stimulating DNA repair mechanisms and immune function to a degree that overcompensates for the radiation damage itself. This "radiation hormesis" has been observed in some animal studies and in ecological studies of natural high-background areas, but the evidence is inconsistent and does not meet the standard required to overturn the LNT model.

2. The DDREF (Dose and Dose Rate Effectiveness Factor): The LSS data reflect acute, high-dose-rate exposures. For chronic, low-dose-rate exposure (which is the relevant scenario for most occupational and environmental exposures), biological repair mechanisms are more effective. The ICRP applies a DDREF of 1.5–2 to reduce the risk coefficient derived from the LSS when applying it to low-dose-rate situations — implicitly acknowledging that LNT overestimates risk at low dose rates.

3. Adaptive response: Cells pre-exposed to a small "priming" dose of radiation show increased resistance to subsequent larger doses, suggesting that low-dose radiation can activate protective mechanisms. The adaptive response has been demonstrated in cell culture and animal experiments, but its relevance to whole-body human exposures is unclear.

4. Natural high-background areas: Studies of populations in Ramsar (Iran), Kerala (India), Yangjiang (China), and Guarapari (Brazil) — where natural background doses are 5–50 times the global average — have generally not found significant increases in cancer rates. However, these studies have limited statistical power due to small populations, migration, and confounding factors.

5. The fundamental statistical problem: At doses below about 100 mSv, the expected excess cancer rate (using LNT: $< 0.5\%$) is far smaller than the statistical fluctuation in the baseline cancer rate (~25%). With realistic epidemiological study sizes ($10^4$–$10^5$ subjects), it is statistically impossible to detect or exclude an excess this small. This is not a failure of experimental design — it is a fundamental limitation.

📊 The numbers illustrate the problem: Suppose we want to detect a 0.5% excess cancer risk (corresponding to 100 mSv under LNT) against a 25% baseline rate. The required study size, for 80% statistical power, is approximately:

$$n \approx \frac{(z_\alpha + z_\beta)^2 \, p(1-p)}{\Delta p^2} \approx \frac{(1.96 + 0.84)^2 \times 0.25 \times 0.75}{(0.005)^2} \approx 5.9 \times 10^6$$

We would need nearly 6 million subjects in both exposed and control groups — and that assumes perfect dose reconstruction and no confounding. No such study exists or is feasible.

29.5.4 Where the Field Stands

The LNT debate is not a conflict between good science and bad science. It is a genuine scientific uncertainty arising from the intersection of two facts:

1. The radiation doses most people receive are too low to produce measurable excess cancer against the high baseline rate.
2. No biological mechanism conclusively proves or disproves a threshold.

The major radiation protection bodies maintain the LNT model as the operational basis for protection, while acknowledging the uncertainty. The ICRP (2007) states explicitly: "The LNT model is the best practical approach for managing radiation risk. ... It does not mean that all doses are harmful; it means that, in the absence of proof to the contrary, it is prudent to assume they might be."

From the perspective of nuclear physics, our job is clear: quantify the physics of radiation production and energy deposition as precisely as possible, and present the biological evidence honestly. The policy decision of how much risk is acceptable at what cost is a social choice, not a physics calculation.

29.6 Radiation Protection: Principles and Practice

29.6.1 The Three Principles

The international system of radiation protection, developed by the ICRP and adopted by national regulators worldwide, rests on three principles:

1. Justification: Any practice involving radiation exposure must produce sufficient benefit to offset the radiation detriment. A medical CT scan is justified when the diagnostic benefit exceeds the radiation risk. An unnecessary repeat scan is not.

2. Optimization (ALARA): All justified exposures should be kept As Low As Reasonably Achievable, taking into account economic and social factors. This is the practical heart of radiation protection — not zero dose (which is impossible), but the lowest dose compatible with the purpose of the activity.

3. Dose limitation: Individual doses must not exceed prescribed limits, even if the practice is justified and optimized. Limits apply to planned (non-emergency) situations and exclude medical exposures to the patient (which are governed by justification and optimization alone).

29.6.2 Dose Quantities: A Hierarchy

The radiation protection dose quantities form a hierarchy designed to account for the different biological effectiveness of different radiation types and the different radiosensitivities of different tissues:

Absorbed dose $D$ (gray, Gy): $$D = \frac{dE}{dm}$$ The fundamental physical quantity — energy deposited per unit mass. Measurable and unambiguous.

Equivalent dose $H_T$ (sievert, Sv): $$H_T = \sum_R w_R \, D_{T,R}$$ where $D_{T,R}$ is the absorbed dose in tissue $T$ from radiation type $R$, and $w_R$ is the radiation weighting factor:

The factor of 20 for alpha particles reflects their high LET and the correspondingly dense ionization damage. An alpha particle depositing 1 mGy in lung tissue produces the same equivalent dose as 20 mGy of gamma radiation.

Effective dose $E$ (sievert, Sv): $$E = \sum_T w_T \, H_T$$ where $w_T$ is the tissue weighting factor reflecting the relative radiosensitivity of each tissue:

💡 Why this matters for radon: When ${}^{218}\text{Po}$ and ${}^{214}\text{Po}$ alpha particles deposit energy in the bronchial epithelium, the absorbed dose (Gy) is multiplied by $w_R = 20$ (for alphas) and then by $w_T = 0.12$ (for lung) to obtain the contribution to the effective dose. The high $w_R$ for alphas is the main reason that radon — despite its modest absorbed dose — dominates the effective dose budget.

29.6.3 Dose Limits

The ICRP-recommended dose limits (Publication 103, 2007):

These limits apply to planned exposure situations and exclude medical exposure (which has no dose limit — the limit is replaced by the justification/optimization principles) and natural background (which cannot be controlled).

📊 Context: The occupational limit of 20 mSv/yr implies an additional lifetime cancer risk (assuming 40-year career) of approximately $40 \times 0.020\,\text{Sv} \times 0.05\,\text{Sv}^{-1} = 4\%$ — roughly comparable to the risk from other accepted occupational hazards. In practice, most radiation workers receive doses well below the limit: the average occupational dose in the US nuclear industry is approximately 1–2 mSv/yr.

29.6.4 Practical Protection: Time, Distance, Shielding

For external radiation sources, the three fundamental protection strategies are:

Time: Dose is proportional to exposure time. Reducing the time spent near a source proportionally reduces the dose.

Distance: For a point source, the dose rate falls as the inverse square of distance: $$\dot{D}(r) = \frac{\dot{D}_0}{r^2}$$ Doubling the distance from a source reduces the dose rate by a factor of 4.

Shielding: Interposing material between the source and the individual attenuates the radiation. The required shielding depends on the radiation type:

• Alpha particles: Stopped by a sheet of paper or a few centimeters of air. Shielding is trivial; the hazard is entirely from internal exposure (inhalation, ingestion).
• Beta particles: Stopped by a few mm of aluminum or plastic. High-$Z$ shielding (lead) should be avoided because it produces bremsstrahlung X-rays.
• Gamma rays/X-rays: Attenuated exponentially by dense, high-$Z$ materials. The half-value layer (HVL) — the thickness that reduces intensity by 50% — is a key design parameter:

• Neutrons: Moderated (slowed) by hydrogen-rich materials (water, polyethylene, concrete), then captured. Borated polyethylene (containing ${}^{10}\text{B}$, a strong neutron absorber) is a standard neutron shielding material.

🔗 Chapter 16 Connection: The half-value layer is directly related to the total attenuation coefficient $\mu$ introduced in Chapter 16: $\text{HVL} = \ln 2 / \mu$. The exponential attenuation law $I = I_0 e^{-\mu x}$ governs the design of all gamma-ray shielding, from hospital X-ray rooms to spent fuel casks. For ${}^{208}\text{Pb}$, the dominant interaction in the 0.5–5 MeV range is Compton scattering, with photoelectric absorption important below ~0.5 MeV and pair production above ~5 MeV.

29.7 Dosimetry: Measuring What Matters

29.7.1 Personal Dosimeters

Radiation workers wear personal dosimeters to record their cumulative dose. The principal technologies are:

Thermoluminescent dosimeters (TLDs): Crystalline materials (LiF:Mg,Ti is the standard; CaSO$_4$:Dy for high sensitivity) that store radiation energy in metastable electron traps. When heated, the trapped electrons are released and emit light proportional to the absorbed dose. TLDs are reusable, measure doses from 0.01 mSv to 10 Sv, and integrate the dose over the wearing period (typically 1–3 months). The physics: radiation creates electron-hole pairs in the crystal; some electrons are trapped in lattice defects; heating provides the thermal energy to de-trap them, and the resulting recombination produces a photon.

Optically stimulated luminescence (OSL) dosimeters: Similar principle, but trapped electrons are released by optical stimulation (green light from an LED) rather than heat. The standard material is Al$_2$O$_3$:C (carbon-doped aluminum oxide). OSL has largely replaced TLD in many programs because it allows multiple re-reads (the optical stimulation depletes only a small fraction of the trapped charge per reading) and has a wider dynamic range.

Electronic personal dosimeters (EPDs): Silicon semiconductor detectors that provide real-time dose and dose-rate readings, with programmable alarms for dose and dose-rate thresholds. Essential for work in high-dose-rate environments (reactor maintenance, medical procedures, emergencies) where immediate feedback is needed.

Film badges: The original personal dosimeter — photographic film whose optical density after development is proportional to dose. Largely obsolete, replaced by TLD and OSL, but historically important and still used in a few countries.

💡 Why LiF is ideal for dosimetry: The effective atomic number of LiF ($Z_{\text{eff}} \approx 8.2$) is close to that of soft tissue ($Z_{\text{eff}} \approx 7.4$). This means the energy absorption per unit mass in LiF closely matches that in tissue across a wide energy range — the dosimeter "sees" the same radiation field that the body does. Materials with higher $Z_{\text{eff}}$ (like CaSO$_4$, $Z_{\text{eff}} \approx 15.3$) over-respond to low-energy photons (where the photoelectric effect, which scales as $Z^{4-5}$, dominates).

29.7.2 Environmental Monitoring

Environmental radiation monitoring serves two purposes: (1) verifying that nuclear facilities operate within regulatory limits, and (2) detecting and tracking radiological emergencies.

Monitoring networks: Most countries operate networks of fixed gamma-ray monitoring stations. The US EPA's RadNet system comprises approximately 140 stations that continuously measure gamma radiation, airborne radioactivity (particulate filters and charcoal cartridges for radioiodine), and precipitation. The European EURDEP (European Radiological Data Exchange Platform) connects over 5,500 monitoring stations across 39 countries — developed after Chernobyl, when the lack of real-time cross-border data sharing delayed the emergency response.

Emergency response instruments: - High-pressure ionization chambers for ambient gamma dose rate (nSv/hr to mSv/hr) - Portable gamma spectrometers (NaI or HPGe) for nuclide identification - Alpha/beta air samplers for airborne contamination - Aerial survey systems (helicopter-mounted NaI arrays) for mapping ground contamination over large areas

Biological dosimetry: When personal dosimeters are not available (as in a radiological emergency or accidental exposure), the absorbed dose can be estimated retrospectively from biological markers: - Dicentric chromosome assay: Radiation-induced aberrations in lymphocytes are scored under the microscope. The dicentric frequency increases with dose following a linear-quadratic relationship: $Y = c + \alpha D + \beta D^2$. This is the "gold standard" of biological dosimetry, accurate to $\pm 0.2$–0.5 Gy. - Lymphocyte depletion kinetics: The rate of decline in peripheral lymphocyte count in the first 48 hours after exposure correlates with dose. A rapid drop to below 1,000/mm$^3$ within 24 hours indicates a dose > 2 Gy. - Electron paramagnetic resonance (EPR) of tooth enamel: Radiation creates stable free radicals in hydroxyapatite that can be detected by EPR spectrometry. Used for retrospective dosimetry of populations exposed years earlier (e.g., near the Semipalatinsk test site).

29.8 The Complete Radiation Budget: Putting It All Together

We can now assemble the complete radiation budget for a "typical" person. Of course, no person is truly typical — individual doses vary enormously depending on where you live, whether you smoke, how many CT scans you've had, and your occupation.

29.8.1 The Average American (NCRP 160)

The pie chart of doses for the average American (circa 2006):

29.8.2 The Global Average (UNSCEAR 2008)

The global average is substantially lower, primarily because medical radiation use is much lower outside the US:

The difference between 6.2 mSv (US) and 3.0 mSv (global) is almost entirely due to higher medical imaging utilization in the US.

29.8.3 Dose Comparisons: Building Intuition

A selection of doses for comparison, spanning seven orders of magnitude:

This table — which your code exercise will generate automatically — is one of the most useful tools for putting radiation risks in perspective.

29.9 Summary

This chapter has surveyed the radiation environment — natural and man-made — and the framework for understanding and managing radiation risk.

1. Natural background radiation is dominated by radon inhalation (~37% of US total), followed by terrestrial sources and cosmic rays. Every human body contains approximately 8,700 Bq of radioactivity, primarily from ${}^{40}\text{K}$ and ${}^{14}\text{C}$.

2. Man-made sources are now dominated by medical imaging, particularly CT scanning (~24% of US total). Nuclear weapons fallout has decayed to negligible levels globally, and normal nuclear power plant operations contribute less than 0.01 mSv/yr to the nearest residents.

3. Deterministic effects (radiation sickness, burns, death) require doses above ~500 mSv and are characterized by a threshold and dose-dependent severity. Acute radiation syndrome progresses through hematopoietic, gastrointestinal, and neurovascular stages with increasing dose.

4. Stochastic effects (cancer) have no proven threshold. The excess cancer risk is approximately 5% per Sv, based primarily on the Life Span Study of atomic bomb survivors.

5. The LNT model — cancer risk proportional to dose with no threshold — is the regulatory standard but remains scientifically unproven below ~100 mSv, where the expected signal is smaller than statistical noise. The hormesis hypothesis, DDREF, and adaptive response provide evidence that LNT may overestimate risk at low doses, but none is conclusive.

6. Radiation protection rests on justification, optimization (ALARA), and dose limitation. The occupational dose limit is 20 mSv/yr averaged over 5 years; the public limit is 1 mSv/yr above background.

7. Dosimetry technologies — TLD, OSL, electronic dosimeters — enable precise tracking of individual doses, while environmental monitoring networks provide early warning of radiological events.

The threshold concept of this chapter: the LNT debate is not resolvable with current data because the expected cancer excess at low doses is far smaller than the statistical uncertainty in the baseline cancer rate. This is an epistemic limitation, not a failure of science — and recognizing it is the beginning of honest risk communication.

Key Equations Summary

What's Next

In Chapter 30, we turn to the instruments of nuclear science itself: accelerators and experimental techniques. Cyclotrons, synchrotrons, linacs, and radioactive beam facilities — from the tabletop to the kilometer scale — are the tools that have made every discovery described in this textbook. We will examine how they work, what they measure, and where the next generation of nuclear physics experiments will take us.