Why Do We Fart, and Why Does It Smell So Bad?

I have spent an embarrassing number of hours reading peer-reviewed research about flatulence, and I want to tell you upfront what surprised me most: almost everything you release is completely odorless. The gases that make a room clear out account for less than 1% of the total volume. The other 99% is chemically boring.

That single fact reframes the whole question. We are not really asking "why do people fart." We are asking two separate questions that have two separate answers. Why does gas accumulate in the gut at all? And why does a tiny sulfurous fraction of it smell like something died?

Below, I break down both questions across nine points, using data from the 2026 CSIRO study in JAMA Network Open, the University of Maryland "Smart Underwear" trial, and the classic 1998 Gut paper that finally identified the specific molecules responsible for the smell. I have included the actual numbers, the actual researchers, and the actual chemistry, because the internet is full of confident guesses on this topic and very little sourcing.

Quick note before we start: this is educational content, not medical advice. If your symptoms have changed suddenly, talk to a clinician.

---

1. We Fart Because Two Completely Different Processes Fill the Gut With Gas

Intestinal gas has exactly two origins, and they are unrelated to each other.

The first is aerophagia, the medical term for swallowed air. Every time we eat, drink, chew gum, smoke, or talk while swallowing, small pockets of atmospheric air go down with it. That air is mostly nitrogen and oxygen. It is odorless. It either comes back up as a belch or continues south.

The second source is colonic fermentation. Your small intestine cannot absorb every carbohydrate you eat. Whatever slips through arrives in the colon, where trillions of resident bacteria treat it as lunch. As they metabolize those leftovers, they release hydrogen, carbon dioxide, and in some people methane.

The second source dominates by a wide margin. In a landmark study at the Royal Hallamshire Hospital in Sheffield, researchers used 24-hour rectal catheter collections in 10 healthy volunteers eating their normal diet plus 200 grams of baked beans. Median daily hydrogen output was 361 mL, with a range from 42 mL all the way to 1,060 mL. When the same volunteers switched to a fiber-free liquid diet for 48 hours, hydrogen production was practically eradicated and total volume dropped to a median of 214 mL per 24 hours.

That is the cleanest demonstration I know of that fermentation, not swallowed air, drives most of what we produce.

2. The "Normal" Number Is Higher Than We Were Told: 5 vs. 14 vs. 32 Per Day

For decades, medical literature cited roughly 14 (±6) daily flatus events. The NIH's MedlinePlus still lists 13 to 21 times per day. Harvard Health describes it as at least 14 times daily, producing between half a liter and a full liter of gas.

Then two studies published within months of each other blew that consensus apart in opposite directions.

Why the enormous gap? Because they measured different things. The CSIRO team measured perceived episodes, meaning what people noticed and remembered to log. The Maryland team measured actual hydrogen release, including events during sleep that nobody could possibly self-report.

The CSIRO study, led by Emily Brindal, ran from November 2024 to February 2025 and found men averaged 5.2 daily episodes versus 4.8 for women, with peaks in the morning, after lunch, and near bedtime. The authors framed their range as "a starting point for conversations about excess."

On the Maryland side, microbiologist Brantley Hall put the problem bluntly: "We don't actually know what normal flatus production looks like." His team's device, described by first author Santiago Botasini, is a hydrogen sensor about the diameter of a quarter that clips onto ordinary underwear.

My takeaway: if you consciously notice yourself passing gas two to seven times a day, you are statistically ordinary. If a sensor were strapped to you, the real count would be several times higher, and that would also be ordinary.

3. Over 99% of What You Release Has No Smell At All

Here is the composition, and it is genuinely unremarkable:

• Nitrogen (N₂) from swallowed air
• Oxygen (O₂), also swallowed, mostly absorbed before it exits
• Carbon dioxide (CO₂) from bacterial fermentation and from stomach acid neutralization
• Hydrogen (H₂), the single largest fermentation product
• Methane (CH₄), but only in some people, which I will get to shortly

Not one of these molecules has an odor. In the Sheffield collections, median CO₂ was 68 mL per 24 hours and unidentified gas, presumably nitrogen, contributed a median of 213 mL. Total daily volume ranged from 476 mL to 1,491 mL, with a median of 705 mL.

So roughly three quarters of a liter per day of essentially odorless gas. As gastroenterologist Dr. Pornchai Leelasinjaroen puts it, "Releasing that gas is simply the body's way of relieving pressure."

4. The Smell Comes From Three Sulfur Gases, and One of Them Dominates

This is the part that took real science to pin down, and the definitive answer came from Fabrizis Suarez, John Springfield, and Michael Levitt at the Minneapolis VA Medical Center, published in Gut in 1998.

Their method was unglamorous and rigorous. They collected flatus quantitatively via rectal tube from 16 healthy subjects who had eaten pinto beans and lactulose to boost output. Each sample was analyzed by gas chromatography and independently rated by two judges trained to detect sulfur compounds by smell.

The measured concentrations:

• Hydrogen sulfide (H₂S): 1.06 µmol/L (SEM 0.2). The rotten-egg note.
• Methanethiol (methyl mercaptan): 0.21 µmol/L (SEM 0.04). Rotting cabbage, decaying vegetation.
• Dimethyl sulfide: 0.08 µmol/L (SEM 0.01). A sweeter, oddly food-like note.

Malodor correlated with hydrogen sulfide concentration at p ≤ 0.001. Hydrogen sulfide is roughly five times more abundant than methanethiol and thirteen times more abundant than dimethyl sulfide, and it is the single best predictor of how bad a given passage smells.

The authors' conclusion was carefully hedged: "Sulphur-containing gases are the major, but not the only, malodorous components of human flatus." When they treated samples with zinc acetate, which binds sulfhydryl compounds, sulfur gas content dropped but odor was not fully eliminated. Something else is contributing, likely short-chain fatty acids and skatole. Activated charcoal, by contrast, removed virtually all odor.

5. Our Noses Are Absurdly Well-Tuned to Hydrogen Sulfide

Think about the numbers above. Those sulfur compounds are present at micromolar concentrations in a gas mixture that also contains hundreds of milliliters of hydrogen and nitrogen. Sulfur gases make up under 1% of the total volume, and hydrogen sulfide is a fraction of that 1%.

And yet you can detect it across a room instantly.

The human olfactory system is far more sensitive to volatile sulfur compounds than to almost anything else. That sensitivity is not an accident. Hydrogen sulfide signals anaerobic decay and it is genuinely toxic at higher concentrations. A nose that detects it at vanishingly small amounts is a nose attached to an ancestor who survived. The disgust reflex is doing exactly what it evolved to do, which is unfortunate for you in an elevator.

6. Methanogens vs. Sulfate Reducers: The Microbial Fork in the Road

This is the mechanism most articles skip, and it explains why some people are simply smellier than others.

Colonic fermentation generates enormous volumes of hydrogen. Hydrogen buildup would stall fermentation, so hydrogenotrophic microbes (hydrogen eaters) consume it. There are three competing routes:

1. Methanogens convert it to methane: 4H₂ + CO₂ → CH₄ + 2H₂O. Odorless.
2. Reductive acetogens convert it to acetate: 4H₂ + 2CO₂ → CH₃COOH + 2H₂O. Odorless.
3. Sulfate-reducing bacteria (SRB) convert it to hydrogen sulfide: 4H₂ + SO₄²⁻ + 2H⁺ → H₂S + 4H₂O. This is the smelly one.

Prevalence matters enormously here. Gut methanogens are found in roughly 30% of people in Western populations and up to about 80% in some African populations, with Methanobrevibacter smithii dominant. Sulfate-reducing bacteria colonize approximately 50% of humans, and Desulfovibrio piger is the most common species. Desulfovibrio species account for around 66% of all colonic SRB.

Critically, methanogens and SRB compete for the same hydrogen, and SRB are the more efficient scavengers because they can operate at lower hydrogen thresholds. In many people, one group effectively dominates. If your colon routes hydrogen toward methane, your output is voluminous but comparatively mild. If it routes hydrogen toward sulfide, the volume may be identical and the experience for everyone nearby is not.

You did not choose this. It was largely settled by your microbiome.

7. Volume vs. Smell: The Foods Responsible Are Not the Same Foods

This distinction is the single most useful practical insight in this entire article.

Foods that increase volume are fermentable carbohydrates, the FODMAP group (Fermentable Oligosaccharides, Disaccharides, Monosaccharides and Polyols).

• Beans and legumes contain raffinose and stachyose, galactooligosaccharides (GOS) linked by alpha-galactosidic bonds. As Monash University's FODMAP team explains, "Because all humans naturally lack this enzyme, GOS is universally malabsorbed." Every single person ferments beans. There is no exception.
• Dairy, if you lack sufficient lactase. Global lactose malabsorption prevalence sits around 68%, so this is the majority condition worldwide, not a rare disorder.
• Cruciferous vegetables, whole grains, onions, and fructose-sweetened drinks.

Foods that increase smell are sulfur-rich, supplying substrate to sulfate-reducing bacteria: red meat and high-protein foods, eggs, garlic, onions, and cruciferous vegetables like broccoli, cauliflower, and Brussels sprouts (which land on both lists).

Gastroenterologist Dr. Anish Sheth notes that for people producing large amounts of foul gas, "if you're eating a super high fiber diet, that could be part of it."

The tradeoff is real and worth stating plainly: a high-fiber, plant-heavy diet produces more gas but generally less offensive gas. A low-fiber, meat-heavy diet produces less volume and worse odor. Frequent farting is usually a sign of a well-fed microbiome, not a broken gut.

8. What Actually Reduces Odor, According to Controlled Trials

Not folk remedies. Things that have been measured.

Activated charcoal. In the 1998 Gut study, activated charcoal removed virtually all odor from collected samples. The researchers also tested a charcoal-lined cushion using gas-tight Mylar pantaloons on eight subjects, and it adsorbed more than 90% of sulfur gases. Charcoal-lined underwear products exist and are built on exactly this finding.

Alpha-galactosidase (the enzyme in Beano). In a randomized, double-blind, placebo-controlled trial, eight healthy volunteers ate a test meal containing 420 grams of cooked beans. A 1,200 GalU dose significantly reduced both breath hydrogen excretion and flatulence severity. This targets volume, not smell.

Bismuth subsalicylate. Suarez and Levitt's team published a separate 1998 Gastroenterology paper showing it markedly decreases hydrogen sulfide release in the human colon. Effective, but not something to take continuously without medical guidance.

Slowing down while eating. Reduces aerophagia. Free, and it targets the swallowed-air fraction that no enzyme can touch.

A structured low-FODMAP trial. Best done with a dietitian, since long-term blanket fiber restriction has real downsides for the microbiome.

9. When I Would Actually See a Doctor

Farting is not the warning sign. In fact, the reverse is more alarming: a sudden inability to pass gas, especially alongside abdominal pain and bloating, can signal a bowel obstruction and warrants urgent attention.

Worth getting evaluated if you notice:

• A sudden, sustained change from your personal baseline
• Gas accompanied by persistent pain, bloating, or altered bowel habits
• Blood in stool, unintentional weight loss, or fever
• Symptoms that consistently follow specific foods, which may point to lactose intolerance, celiac disease, IBS, or SIBO

Northwell gastroenterologist Aditya Sreenivasan said he expects to use the new CSIRO figures to give concerned patients "a benchmark for what is considered 'normal.'" That is genuinely useful. Most people worried about their gas are within the normal range and have simply never had a number to compare against.

---

What I Took Away From All This

We fart because bacteria in the colon are doing productive metabolic work on food our own enzymes cannot break down, and because we swallow air. The volume is mostly hydrogen, carbon dioxide, and nitrogen, and none of that smells.

The odor is a rounding error by volume: hydrogen sulfide at around 1.06 µmol/L, plus methanethiol and dimethyl sulfide, produced by sulfate-reducing bacteria competing with methanogens for the same hydrogen. Whether your colon sends hydrogen down the methane path or the sulfide path is largely a matter of which microbes colonized you, which is why two people eating identical meals can have dramatically different results.

And the numbers are more forgiving than most people assume. Somewhere between 5 self-noticed episodes and 32 sensor-detected ones per day, producing roughly 500 to 1,500 mL of gas, puts you squarely in the range of a healthy, well-functioning digestive system.

---

Sources

• Suarez FL, Springfield J, Levitt MD. "Identification of gases responsible for the odour of human flatus and evaluation of a device purported to reduce this odour." Gut 1998;43(1):100-104. doi:10.1136/gut.43.1.100
• Brindal E, et al. "Regular Flatulence Patterns Among Community-Dwelling Individuals in Australia." JAMA Network Open 2026. doi:10.1001/jamanetworkopen.2026.15637
• Botasini S, et al. "Smart underwear: A novel wearable for long-term monitoring of gut microbial gas production via flatus." Biosensors and Bioelectronics: X 2025;27:100699. doi:10.1016/j.biosx.2025.100699
• Tomlin J, Lewis C, Read NW. "Investigation of normal flatus production in healthy volunteers." Gut 1991. PMID: 1648028
• Suarez FL, Furne JK, Springfield J, Levitt MD. "Bismuth subsalicylate markedly decreases hydrogen sulfide release in the human colon." Gastroenterology 1998;114(5):923-929
• Di Stefano M, et al. "The effect of oral alpha-galactosidase on intestinal gas production and gas-related symptoms." PMID: 17151807
• Rey FE, et al. "Metabolic niche of a prominent sulfate-reducing human gut bacterium." PNAS 2013;110(33):13582
• MedlinePlus (NIH), "Gas"; Harvard Health, "Gas (Flatulence)"; Monash University FODMAP program