On this page

FACTORS 2026.10 · 2026-10-06

Methodology

Every number in Touch Grass is an estimate. This page shows how each one is made, what it is compared with, how sure we are and where it comes from.

ACTIONS

SOURCES

REGIONS

Try it

Tap the mark to see how a number is made

Every figure in the app carries this mark. It opens the formula, the comparison, the likely range and the source, just as the table below lists them.

Walked or cycled instead of driving

5 km

approximately 1.1 kg

What the numbers are

Estimates of CO2e avoided against a stated alternative, good for scale and ranking.

Every figure in Touch Grass is an estimate of the greenhouse gases you avoided compared with a stated alternative, called the counterfactual. The alternative is assumed, never observed, so the numbers are indicative. They are good for ranking actions and showing rough scale. They are not carbon accounting.

The unit is CO₂e, carbon dioxide equivalent: all greenhouse gases expressed as the amount of CO₂ with the same warming effect over 100 years. Each factor comes from a published source, listed in the table at the bottom of this page.

estimated kg CO₂e avoided = quantity × kg CO₂e avoided per unit

The comparison behind every number

A figure only means something next to its comparison. "Walked instead of driving" is compared with driving an average car, which emits about 0.21 kg CO₂e per kilometre in the UK government factors, including fuel production. If you would not have driven, there is nothing to avoid, so please do not log it.

The comparison is written next to every figure in the app ("vs driving an average car") and in full in the table below, under "Compared with".

Ranges and confidence

Every factor carries a confidence level and a plausible range.

High
An official or peer-reviewed factor and a simple, well-defined comparison.
Typically within about ±30%.
Medium
Good sources, but the answer depends on your circumstances: your vehicle, your grid, your diet, what you would otherwise have done.
Often ±50%.
Low
Assumption-heavy, context-dependent, or the sources disagree. For some people the true figure could be zero or negative.
Could be out by a factor of two or more.
Not quantified
No credible CO2e figure exists. The action earns XP for showing up, and Touch Grass never prints a number for it.
No number is shown.

Each action has a central value, a low and a high value, and a confidence level. The ranges are published where the source gives one and judged where it does not. The app shows one central number, rounded to two significant figures, and the range when you open the details.

Where you live changes the number

Electricity, car and public-transport factors adjust to one of 45 regions.

One kilowatt-hour of electricity does not always mean the same CO₂. Touch Grass holds the average carbon intensity of the grid for 45 countries and regions, from Ember's yearly data on a life-cycle basis. The default, "World average", is about 458 g CO₂e per kWh. France is about 41 g and India about 671 g (2025).

kg per unit = non-grid part + kWh per unit × grid kg CO₂e per kWh of your region

Car, bus and rail factors also change by region where we have regional values: the average car is 0.2099 kg CO₂e per km by default, 0.2442 for the United States and 0.14 for India. Heating and hot-water actions depend on how your home is heated. A heat pump is modelled with a coefficient of performance of 3, and gas at 0.213 kg CO₂e per kWh.

Actions about food, waste and stuff do not depend on the grid and are the same everywhere. Annual grid averages are used, not hourly values, and a logged value is stored at the factor version in force on the day, so a later update never rewrites your history.

Every factor

One row for every action you can log: the value, how far it could be off, what it is compared with and where it comes from.

51 actions. Every value is an estimate in kg CO₂e avoided per unit, on the world-average grid.

Emission factors: one row for every action you can log
Action	Estimate	Likely range	Compared with	Sources
Walked or cycled instead of driving MoveHigh	approximately 0.21 kg CO₂e per km Varies by region	0.14 to 0.34 kg	the same trip in an average car Show workingHide working The same trip driven in an average car (UK fleet average, unknown fuel) with you as the driver. Counts fuel burned plus fuel production ("well-to-wheel"); excludes car manufacture. km x (tailpipe 0.16591 + fuel production 0.04399) = km x 0.2099 kgCO2e. Low = small hybrid 0.1441; high = large petrol car 0.3413. Only log trips that would otherwise have been driven. US fleet average is higher: 0.244 kgCO2e/km tailpipe only (EPA, 22.8 mpg). India: a petrol hatchback is about 0.14 and a scooter about 0.0368 kgCO2/km (India GHG Program 2015, tailpipe CO2 only) - use the user's real vehicle. Cross-check: Wynes & Nicholas (2017) put living car-free at 2.4 tCO2e/yr.	2026UK Department for Energy Security and Net Zero 2026US Environmental Protection Agency 2015India GHG Program 2017Environmental Research Letters
Took the bus instead of driving MoveMedium	approximately 82 g CO₂e per km Varies by region	53 g to 0.16 kg	driving it alone, minus the bus’s own emissions Show workingHide working The same distance driven alone in an average car, compared with the average emissions per passenger-km of a local bus at typical UK occupancy. km x (car 0.2099 - average local bus 0.128) = km x 0.0819. Low uses a local bus outside London (0.1573); high uses a coach (0.046). Bus emissions per passenger depend almost entirely on how full the bus is. US average transit bus: 0.041 kgCO2/passenger-km (EPA 2025); Indian city bus: 0.015161 kgCO2/passenger-km (India GHG Program 2015). If the alternative would have been a motorbike or a shared car the saving is smaller and can be zero.	2026UK Department for Energy Security and Net Zero 2025US Environmental Protection Agency 2015India GHG Program
Took the train or metro instead of driving MoveHigh	approximately 0.17 kg CO₂e per km Varies by region	0.13 to 0.2 kg	driving it alone, minus the train’s own emissions Show workingHide working The same distance driven alone in an average car, compared with average national rail emissions per passenger-km. km x (car 0.2099 - national rail 0.0399) = km x 0.17. Low uses US commuter rail (0.0826 kgCO2/pkm, EPA 2025); high uses electric international rail (0.0125). Metro and tram are similar or better: London Underground 0.0228, light rail/tram 0.0287 kgCO2e/passenger-km (UK 2026). Indian Railways: about 0.007837 kgCO2/passenger-km (2014-15).	2026UK Department for Energy Security and Net Zero 2025US Environmental Protection Agency 2015India GHG Program
Shared the ride MoveMedium	approximately 0.21 kg CO₂e per km Varies by region	40 g to 0.34 kg	everyone driving alone; your share of the car-km avoided Show workingHide working Each extra passenger would otherwise have driven the same distance alone in an average car. One car-kilometre is avoided per shared kilometre per extra passenger. km x extra passengers x 0.2099 (one average car-km avoided). Low: the passenger would otherwise have taken the train (0.0399); high: a large petrol car is left at home (0.3413). DOUBLE-COUNTING RISK: the saving belongs to the car, not to each person. Log it once (the driver logs the passengers, or each rider logs a 1/n share). Detours to pick people up reduce the saving and are not modelled.	2026UK Department for Energy Security and Net Zero
Rode my e-bike or e-scooter instead of driving MoveHigh	approximately 0.21 kg CO₂e per km Varies by region	0.13 to 0.34 kg	the same trip in an average car, minus charging Show workingHide working The same trip driven alone in an average car. The e-bike uses a little grid electricity (about 0.01 kWh per km). km x (car 0.2099 - 0.01 kWh x grid). At the world-average grid (0.45849 kg/kWh) that is 0.2053 per km. kWhPerUnit is negative because the action USES electricity. E-bike electricity is tiny: McQueen et al. (2019) calculate 3.778-12.568 gCO2 per mile across US grid regions (4.905 in the US Northwest); a 160 Wh pedelec battery giving 25-35 km implies 0.005-0.006 kWh/km (Petrillo et al. 2018). 0.01 kWh/km is our rounded, deliberately cautious assumption. Applies to PRIVATELY OWNED e-bikes and e-scooters. Shared dockless e-scooters are different: Hollingsworth et al. (2019) found 202 gCO2e per passenger-mile over the life cycle (vs 414 for a car) and only 34% of rides replaced a car trip - do not award this factor for shared scooters that replaced walking.	2026UK Department for Energy Security and Net Zero 2019Portland State University, Transportation Research and Education Center 2018IntechOpen 2019Environmental Research Letters 2026Ember
Drove electric instead of petrol MoveMedium	approximately 0.11 kg CO₂e per km Varies by region	71 g to 0.2 kg	the same distance in an average petrol car, minus charging Show workingHide working The same distance in an average petrol car. The electric car draws 0.20358 kWh per km from the grid (UK average battery-electric car). km x (petrol car 0.2075 - 0.20358 kWh x grid). World-average grid: 0.114; India: 0.071; France: 0.199. Operating emissions only: battery and vehicle manufacturing are excluded on both sides (an EV carries a larger manufacturing footprint that is repaid over the first years of driving). Grid losses (5-10%) are not added. The UK Government's own all-scope figure for an average battery-electric car on the 2026 UK grid is 0.04 kgCO2e/km.	2026UK Department for Energy Security and Net Zero 2026Ember
Worked from home instead of commuting by car MoveLow	approximately 3.5 kg CO₂e per day Varies by region	0 to 12 kg	your car commute, minus the extra energy used at home Show workingHide working A return car commute of 29.2 km (England average commuting trip of 14.6 km each way) minus the extra heating and equipment energy used at home over an 8-hour day. 2 x 14.6 km x 0.2099 - 8 h x 0.32393 kgCO2e/h (UK homeworking factor: heating 0.30234 + equipment 0.02159) = 3.54. Low = same distance by train (-1.43, i.e. working from home is WORSE). High = double the distance by car with no extra heating (12.09). Can be NEGATIVE. The app should ask for the real round-trip distance and mode and compute: roundTripKm x modeFactor - hours x 0.324 (heating season) or x 0.022 (no heating or cooling needed). Break-even for a car commuter is about 6.2 km each way, which matches the IEA finding that home-working only cuts CO2 for car commutes longer than about 6 km (IEA text seen only as a search excerpt - site blocks automated access). The UK heating factor is an annual average for gas-heated homes; air-conditioning in hot climates is not covered.	2026UK Department for Energy Security and Net Zero 2026UK Department for Energy Security and Net Zero 2025UK Department for Transport 2020International Energy Agency
Took the train instead of a short flight MoveMedium	approximately 0.11 kg CO₂e per km	57 g to 0.22 kg	an economy seat on a short-haul flight, minus the train Show workingHide working An economy seat on a short-haul flight over the same great-circle distance, including the non-CO2 warming effects of flying at altitude, compared with national rail. km x (short-haul economy flight 0.1483 - national rail 0.0399) = km x 0.1084. Low = flight without non-CO2 effects (0.0968); high = UK domestic flight average (0.2628). UK flight factors include an 8% uplift for indirect routing and a 1.7 multiplier on CO2 for non-CO2 effects ("radiative forcing"); the multiplier is scientifically uncertain. Use the great-circle distance, not the (longer) rail distance. Electric high-speed rail is cleaner than the national-rail average used here (0.0125 for international rail).	2026UK Department for Energy Security and Net Zero 2026UK Department for Energy Security and Net Zero
Took the train instead of a short flight (distance unknown) MoveMedium	approximately 54 kg CO₂e per trip	28 to 110 kg	a one-way short-haul flight of about 500 km, minus the train Show workingHide working One economy seat on a 500 km one-way flight (including non-CO2 effects) compared with the same journey by national rail. 500 km x 0.1084 = 54.2 kgCO2e one way. For a specific route use the per-km action with the real great-circle distance. Illustrative distance. Approximate great-circle distances (our haversine arithmetic): London-Paris 347 km, London-Edinburgh 534 km, San Francisco-Los Angeles 544 km, Sydney-Melbourne 706 km, Delhi-Mumbai 1,137 km. A return trip is two units.	2026UK Department for Energy Security and Net Zero 2026UK Department for Energy Security and Net Zero
Car-free day MoveNot quantified	Not quantified	No number shown	not quantified: the kilograms come from the trips you log
Plant-based meal EatMedium	approximately 1.5 kg CO₂e per meal	0.97 to 2.6 kg	a typical meal with meat Show workingHide working One of three daily meals of a medium meat-eater (50-99 g of meat a day), replaced by a fully plant-based (vegan) meal. (medium meat-eater 7.04 - vegan 2.47 kgCO2e/day) / 3 meals = 1.52. Low uses a low meat-eater (5.37), high a high meat-eater (10.24). Scarborough et al. linked the real diets of 55,504 UK adults to the Poore & Nemecek life-cycle database (farm to retail, global-average production, land-use change included, GWP100), standardised to 2,000 kcal/day. Assuming three equal meals is our simplification. Do not also log a "vegetarian day" for the same day.	2023Nature Food 2018Our World in Data / Science
Plant-based instead of a beef meal EatMedium	approximately 6.3 kg CO₂e per meal	3 to 9.8 kg	the same meal made with about 100 g of beef Show workingHide working A portion of beef supplying 20 g of protein (about 100 g of meat), replaced by tofu supplying the same protein. Per 20 g protein (Poore & Nemecek global means): beef from beef herds 9.98, beef from dairy herds 3.37, tofu 0.4, pulses 0.17 kgCO2e. Central = mean of the two beef types minus tofu = 6.28. Low = dairy-herd beef vs tofu; high = beef-herd beef vs pulses. Beef is the single largest lever in most diets, but its footprint varies enormously by farm and country: the global means are 99.48 (beef herd) and 33.3 (dairy herd) kgCO2e per kg, including land-use change and methane. The central value is an unweighted mean of those two - we could not verify the production split, so treat it as indicative. Many national beef footprints (without land-use change) are lower.	2018Our World in Data / Science 2018Our World in Data / Science 2018Our World in Data / Science
Chicken instead of beef EatMedium	approximately 5.7 kg CO₂e per serving	2.3 to 9 kg	the same 100 g serving as beef Show workingHide working 100 g of beef (retail weight) replaced by 100 g of poultry meat. 0.1 kg x (beef 66.39 [mean of 99.48 and 33.3] - poultry 9.87) = 5.65 kgCO2e. Shows that the choice of meat matters almost as much as going meat-free. Lamb (39.72) and cheese (23.88) are also high; pork is 12.31 kgCO2e/kg (global means).	2018Our World in Data / Science
Vegetarian day EatMedium	approximately 2.9 kg CO₂e per day	1.2 to 6.1 kg	a day of typical meals with meat Show workingHide working A normal day of eating for a medium meat-eater (50-99 g of meat a day), compared with a typical vegetarian day. medium meat-eater 7.04 - vegetarian 4.16 = 2.88 kgCO2e/day. Low: low meat-eater (5.37); high: high meat-eater (10.24). Vegetarian diets that lean heavily on cheese and other dairy save less. Do not also log individual plant-based meals for the same day.	2023Nature Food
Fully plant-based day EatMedium	approximately 4.6 kg CO₂e per day	2.9 to 7.8 kg	a day of typical meals with meat Show workingHide working A normal day of eating for a medium meat-eater, compared with a typical vegan day. medium meat-eater 7.04 - vegan 2.47 = 4.57 kgCO2e/day. Over a year this is 1.67 tCO2e. Wynes & Nicholas (2017) estimate 0.8 tCO2e/yr for a plant-based diet using older, lower food footprints - the two sources bracket the plausible range.	2023Nature Food 2017Environmental Research Letters
Plant milk instead of dairy EatHigh	approximately 2.2 kg CO₂e per litre	2 to 2.4 kg	the same amount of cow’s milk Show workingHide working One litre of dairy milk replaced by one litre of plant milk (average of soy, oat, almond and rice). dairy 3.15 - mean plant milk 0.942 (soy 0.98, oat 0.9, almond 0.7, rice 1.18) = 2.21 kgCO2e per litre. A 200 ml glass is one fifth of this. Almond milk has the lowest carbon footprint of the four but by far the highest water use; oat and soy are good all-rounders.	2018Our World in Data / Science
Saved food from the bin EatMedium	approximately 3.6 kg CO₂e per kg	0.84 to 4.6 kg	binning it and buying the same food again Show workingHide working The food would have been thrown away and replaced by buying the same amount again. Counts the whole supply chain of the wasted food. UK household food waste: 16 MtCO2e / 4.4 Mt edible food = 3.64 kgCO2e per kg (WRAP 2022 data). Cross-checks: UK Government 2026 "food and drink" factor 3.7; EPA WARM v16 source reduction 4.03. Low = EPA non-meat food (0.84); high = EPA mixed food plus avoided landfill methane (4.59). What was saved matters far more than how much: EPA WARM gives 0.49 kgCO2e/kg for fruit and vegetables, 0.73 for bread, 1.93 for dairy and 33.2 for beef. Do not add "composting" for the same food.	2025WRAP 2026UK Department for Energy Security and Net Zero 2023US Environmental Protection Agency
Rescued a meal (leftovers, food-rescue app) EatMedium	approximately 1.5 kg CO₂e per meal	0.35 to 1.9 kg	binning the meal and cooking another Show workingHide working A 420 g meal that would otherwise have been binned and replaced. 0.42 kg (WRAP standard meal) x 3.64 kgCO2e/kg = 1.53 kgCO2e. Same evidence as "food waste avoided" expressed per meal - log one or the other, never both. Commercial food-rescue apps often quote 2.5-2.7 kgCO2e per meal; that is at the top of our range.	2020WRAP 2025WRAP 2023US Environmental Protection Agency
Chose local, seasonal produce EatLow	approximately 70 g CO₂e per meal	0 to 0.1 kg	the same meal with ingredients shipped the usual distances Show workingHide working The same meal made from ingredients transported the usual distances by road and sea. Transport is about 6% of diet emissions (EU diets, Sandstrom et al. 2018 via Our World in Data). 6% x 2.35 kgCO2e (a medium meat-eater's average meal) = 0.14 if ALL transport vanished (high). Central assumes half of it is avoided: 0.07. HONEST VERSION: the climate effect is small - what you eat matters far more than where it came from. Transport is 0.5% of beef's footprint and under 10% for most foods. Local food grown in heated greenhouses or stored for months can be worse than imports. The real exception is AIR-FREIGHTED produce (out-of-season asparagus, green beans, berries): air freight emits 1.13 kgCO2e per tonne-km against 0.023 by sea, so 1 kg flown 8,000 km adds about 9 kgCO2e - but only 0.16% of food miles are flown. Reward this action for its other benefits (freshness, local growers) and show the number as "small".	2020Our World in Data 2020Our World in Data 2018Our World in Data / Science 2023Nature Food
Heating 1 °C lower PowerLow	approximately 0.77 kg CO₂e per day Varies by region	0.5 to 2.2 kg	the same home kept 1 °C warmer today Show workingHide working The same gas-heated home kept 1 degC warmer on a heating-season day. Typical UK home: 11500 kWh gas/yr (Ofgem median) x 82% space heating = 9430 kWh; x 7.000000000000001% per degC (IEA/European Commission) = 660 kWh/yr; / 182 heating days = 3.63 kWh gas per day; x 0.21252 kgCO2e/kWh = 0.77. Low = Ofgem low-use home (7500 kWh). High = Energy Saving Trust's 22->21 degC saving of GBP 120/yr at 6.5p/kWh = 1846 kWh/yr. PER HOUSEHOLD, not per person - one household member logs it, or divide by the number of people. Only valid on days the heating is actually on. Very climate- and building-dependent (the two sources we use differ by a factor of three). heatKWhPerUnit here is kWh of GAS, not useful heat. For electric heating multiply the kWh by your grid intensity instead (and divide by about 3 for a heat pump). Never suggest going below a healthy indoor temperature (WHO recommends at least 18 degC).	2022International Energy Agency / European Commission 2023Ofgem 2026Energy Saving Trust 2026Energy Saving Trust 2026Energy Saving Trust 2026UK Department for Energy Security and Net Zero 2026UK Department for Energy Security and Net Zero
Air-dried a load PowerMedium	approximately 0.69 kg CO₂e per load Varies by region	41 g to 2 kg	the same load in a tumble dryer Show workingHide working The same load dried in an electric tumble dryer (average measured UK cycle: 1.52 kWh). 394 kWh/yr / 260 cycles = 1.515 kWh per measured drying cycle; x world-average grid 0.45849 = 0.69 kgCO2e. Measured UK cycles vary widely: 33% used under 1 kWh, 75% under 2 kWh and 10% over 3 kWh (many are short top-up cycles). A full load in a standard US electric dryer that just meets ENERGY STAR uses about 2.15 kWh (8.45 lb test load / 3.93 lb per kWh); heat-pump dryers use roughly half. Low = 1 kWh on the French grid; high = 3 kWh on the Indian grid. Only counts if you own and would have used a dryer. Wynes & Nicholas (2017): 0.21 tCO2e/yr.	2012UK Government (DEFRA / DECC / Energy Saving Trust) 2014US EPA ENERGY STAR 2026Energy Saving Trust 2026Ember 2017Environmental Research Letters
Switched standby devices off at the wall PowerLow	approximately 0.23 kg CO₂e per day Varies by region	10 g to 0.76 kg	the household’s devices left on standby all day Show workingHide working All the household's TVs, consoles, set-top boxes, chargers and similar devices left on standby for the day. Energy Saving Trust: about GBP 45 a year saved at 24.7p/kWh = 182 kWh/yr = 0.5 kWh per day (an average draw of 21 W); x world-average grid = 0.23 kgCO2e. PER HOUSEHOLD and derived from a money figure, so treat as rough. High = the 47 W average standby load measured in English homes in 2010-11 (1.13 kWh/day) on the Indian grid; today's devices draw far less (EU rules cap standby at 0.5-1 W per device). Low = half the Energy Saving Trust figure on the French grid. Unplugging a single phone charger saves almost nothing.	2026Energy Saving Trust 2026Energy Saving Trust 2012UK Government (DEFRA / DECC / Energy Saving Trust) 2026Ember
Replaced an old bulb with an LED PowerMedium	approximately 33 g CO₂e per bulb day Varies by region	1 to 91 g	a 60 W bulb left in the fitting (day one only is counted) Show workingHide working A 60 W incandescent (or equivalent halogen) bulb left in the same fitting, used the US-average 1.6 hours a day. 60 W x 75% saving (LEDs use at least 75% less than incandescent, US DOE) x 1.6 h/day (US DOE/PNNL average per lamp) = 0.072 kWh per bulb per day; x world-average grid = 0.033 kgCO2e. That is 26.3 kWh a year per bulb. A one-off action with a small ongoing benefit: award XP once, then count the saving passively. If the old bulb was already a compact fluorescent the saving is close to zero. Low = 40 W bulb used 1 h/day on the French grid; high = 60 W bulb used 3 h/day (typical for outdoor lights) on the Indian grid. Energy Saving Trust: LEDs use 80% less electricity than halogens. The "at least 75% less" DOE statement was only seen as a search excerpt (page moved).	2012US Department of Energy / Pacific Northwest National Laboratory 2024US Department of Energy 2026Energy Saving Trust 2026Ember
AC 1 °C warmer, or a fan instead PowerNot quantified	Not quantified	No number shown	not quantified: we have no sourced factor for cooling yet
Shorter shower WaterMedium	approximately 52 g CO₂e per minute Varies by region	31 g to 0.15 kg	the same shower, that many minutes longer Show workingHide working One more minute under a hot shower flowing at 7.6 litres a minute, heated by a gas boiler. 7.57 L/min (2 US gal/min) x 4.186 kJ/kg.K x 25 K / 3600 = 0.22 kWh of heat per minute; / 0.9 boiler efficiency x 0.21252 kgCO2e per kWh gas = 0.052 kgCO2e. Electric water heater: 0.22 kWh x grid (world average: 0.101). Physics-based: flow rate x temperature rise. The 25 K rise (e.g. 15 -> 40 degC) and 90% boiler efficiency are our assumptions. Low = 1.5 gal/min low-flow head, 20 K rise, gas; high = standard 2.5 gal/min head, 30 K rise, electric resistance heater on the world-average grid. heatKWhPerUnit is USEFUL HEAT: divide by 0.9 and multiply by the gas factor, or multiply by the grid intensity for electric heaters (divide by about 3 for a heat pump). Solar water heaters and unheated showers: zero.	2026US Environmental Protection Agency 2023US National Institute of Standards and Technology 2026UK Department for Energy Security and Net Zero 2026Energy Saving Trust 2026Ember
Saved hot water WaterMedium	approximately 7 g CO₂e per litre Varies by region	0 to 19 g	heating that many more litres Show workingHide working One more litre of water heated by 25 degC in a gas boiler, then supplied and treated as waste water. 4.186 kJ/kg.K x 25 K / 3600 = 0.0291 kWh of heat per litre; / 0.9 x 0.21252 = 0.0069 kgCO2e, plus 0.00036 for water supply and treatment (UK 2026) = 0.0072. ENERGY BASIS: almost all of the footprint of water use at home is the energy to HEAT it - about 19 times more than pumping and treating it. Use for baths (a 80-litre bath is 80 units), washing-up and hot taps. Low = heat pump on the French grid; high = electric resistance heater on the Indian grid. Do not also log "shorter shower" for the same water.	2023US National Institute of Standards and Technology 2026UK Department for Energy Security and Net Zero 2026Ember
Washed at 30 °C instead of 40 °C WaterMedium	approximately 0.1 kg CO₂e per load Varies by region	5 g to 0.34 kg	the same load at 40 °C Show workingHide working The same load washed on a 40 degC cycle (average measured UK cycle: 0.585 kWh). Average measured wash cycle 166 kWh/yr / 284 cycles = 0.585 kWh; washing at 30 degC uses 38% less (Which? lab tests) = 0.222 kWh saved per load. kgCO2e = kWh x grid intensity (world average 0.45849: 0.102). Dropping from 60 degC to 30 degC saves at least 0.51 kWh. kWh is the robust part; kgCO2e depends on the grid. Low = an efficient machine saving half as much on a very clean grid (France); high = 60->30 degC on a coal-heavy grid (India). The measured cycle average is from 2010-11 UK machines; modern machines use somewhat less. Wynes & Nicholas (2017) put cold-water washing at 0.25 tCO2e/yr for North American households.	2012UK Government (DEFRA / DECC / Energy Saving Trust) 2026Which? 2026Ember 2017Environmental Research Letters
Washed cold instead of 40 °C WaterMedium	approximately 0.17 kg CO₂e per load Varies by region	8 g to 0.44 kg	the same load at 40 °C Show workingHide working The same load washed on a 40 degC cycle. 0.585 kWh x 62% (Which? lab tests) = 0.362 kWh saved per load; x world-average grid 0.45849 = 0.166 kgCO2e. Log either this or the 30 degC action for a load, not both. Health guidance still recommends an occasional hot wash for towels, bedding and when someone is ill.	2012UK Government (DEFRA / DECC / Energy Saving Trust) 2026Which? 2026Ember
Tap off while brushing WaterMedium	approximately 10 g CO₂e per day	4 to 10 g	the tap left running, morning and evening Show workingHide working The cold tap left running while brushing, morning and evening. 8 US gallons a day (EPA WaterSense) = 30.3 litres x 0.362 g CO2e per litre (UK 2026 water supply + treatment) = 11 g CO2e. HONEST VERSION: about 11 g CO2e a day - negligible for the climate. It saves roughly 30 litres of drinking water, which matters where water is scarce. Show litres saved, not CO2e. High = 9 L/min tap (UK Consumer Council for Water via Energy Saving Trust) for 2 x 2 minutes; the water-supply factor is UK-specific.	2026US Environmental Protection Agency 2026UK Department for Energy Security and Net Zero 2026Energy Saving Trust
Bought a top second-hand StuffLow	approximately 2.9 kg CO₂e per item	2.2 to 7.5 kg	the same garment bought new Show workingHide working A new T-shirt would have been made and bought. WRAP finds that 65% of second-hand purchases really displace a new one. 0.2 kg x 22.31 kgCO2e per kg of new clothing (UK 2026) = 4.46; x 0.646 displacement = 2.88 kgCO2e. Low = 150 g shirt; high = WRAP's "over 7.5 kg" for a cotton T-shirt. Garment weight (200 g) is an assumption. If the purchase is an extra you would not otherwise have bought, the saving is zero - that is what the displacement rate captures on average. Shipping of second-hand items is not deducted.	2026UK Department for Energy Security and Net Zero 2025WRAP
Bought jeans second-hand StuffMedium	approximately 13 kg CO₂e per item	10 to 30 kg	the same jeans bought new Show workingHide working A new pair of jeans would have been made and bought; 65% of second-hand purchases displace a new one (WRAP). Levi's 501 life-cycle assessment: fibre 2.9 + fabric 9 + cut/sew/finish 2.6 + sundries/packaging 1.7 + transport/retail 3.8 = 20 kgCO2e before the customer wears them (of 33.4 over the whole life); x 0.646 = 12.9. High = WRAP's "over 30 kg" claim. Washing and drying the jeans (12.5 kgCO2e over their life in the Levi's study) happens either way and is not counted.	2015Levi Strauss & Co. 2025WRAP 2026UK Department for Energy Security and Net Zero
Repaired instead of replacing StuffLow	approximately 7 kg CO₂e per item	3.7 to 73 kg	buying a new replacement Show workingHide working A new replacement would have been bought. WRAP finds 82% of repairs genuinely prevent a new purchase. Default (small electrical item, 1.5 kg): 1.5 kg x 5.648 kgCO2e/kg (UK 2026) = 8.5; x 0.822 displacement = 7 kgCO2e. See variants for other item classes. Item-class dependent - use the variants. The displacement rate was measured for clothing and is applied to other items by analogy (assumption). Item weights are assumptions. The repair itself (parts, travel) is not deducted.	2025WRAP 2026UK Department for Energy Security and Net Zero 2015Levi Strauss & Co. 2025Apple 2025Apple
Borrowed or rented instead of buying StuffLow	approximately 7.6 kg CO₂e per item	0.36 to 10 kg	buying the item new Show workingHide working Buying the item new for a one-off use. The shared item is assumed to serve about ten borrowers. Default (tool or small appliance, 1.5 kg): 8.5 kgCO2e to make (UK 2026 small electricals 5.648 kgCO2e/kg) x 0.9 = 7.6. Low = a 300 g library book (paper 1.344 kgCO2e/kg); high = a 500 g outfit (clothing 22.31 kgCO2e/kg). ASSUMPTION-DRIVEN: item weights and the "ten borrowers" share are ours, there is no measured displacement rate for lending, and rental delivery or dry-cleaning is not deducted. Only log when you genuinely would have bought the item. Use for XP and storytelling more than for the CO2e total.	2026UK Department for Energy Security and Net Zero
Keeping my phone another year StuffMedium	approximately 15 kg CO₂e per year	11 to 22 kg	replacing it now with a new phone Show workingHide working Replacing the phone now with a new one that would itself be used for three years. iPhone 17 (256 GB): 55 kgCO2e life cycle, of which production 76% + transport 4% = 44 kgCO2e before first use. Delaying replacement by a year avoids one year's share: 44 / 3 = 14.7. Low = 4-year replacement cycle; high = 2-year cycle. A once-a-year action, not a daily one - award it once per year. One manufacturer's own figure for one model; other phones range from roughly 40 to 90 kgCO2e. Charging a phone for a year is only a few kgCO2e, so keeping the old one is almost pure gain. A laptop is worth about 22 kgCO2e per extra year (MacBook Air M4: 120 kgCO2e, 74% before first use, 4-year life).	2025Apple
Sold, donated or swapped something StuffNot quantified	Not quantified	No number shown	not quantified: the saving belongs to whoever uses it next
Recycled aluminium cans WasteMedium	approximately 0.13 kg CO₂e per item	81 g to 0.18 kg	landfill, and new cans made from ore Show workingHide working The can goes to landfill and new cans are made from virgin aluminium instead of recycled metal. Per kg: EPA WARM v16 (recycling -9.13 vs landfill 0.02 MTCO2e/short ton) = 10.09 kgCO2e/kg; UK 2026 (primary 9.114 vs recycled 0.994 kgCO2e/kg) = 8.12. Mean 9.11 x 0.014 kg per can = 0.127. Aluminium is the star of recycling - recycled metal needs a small fraction of the energy of smelting new. The per-kg factor is well evidenced; the 14 g can weight is an UNVERIFIED assumption (range used: 10-17.5 g). Only counts where cans are actually collected and recycled.	2023US Environmental Protection Agency 2026UK Department for Energy Security and Net Zero
Recycled glass bottles or jars WasteMedium	approximately 91 g CO₂e per item	66 g to 0.29 kg	landfill, and new glass from raw materials Show workingHide working The bottle goes to landfill and new glass is made from virgin raw materials. Per kg: EPA WARM v16 0.33; UK 2026 (primary 1.403 vs recycled 0.823) 0.58 kgCO2e/kg. Mean 0.46 x 0.2 kg = 0.091. High = a 500 g wine bottle at the UK factor. Small per item because glass is heavy but not carbon-intensive per kg. Refilling or reusing a bottle beats recycling it. The 200 g bottle weight is an assumption.	2023US Environmental Protection Agency 2026UK Department for Energy Security and Net Zero
Recycled plastic bottles WasteMedium	approximately 28 g CO₂e per item	11 to 41 g	landfill, and new plastic from oil Show workingHide working The bottle goes to landfill and new bottles are made from virgin PET. Per kg PET: EPA WARM v16 1.17; UK 2026 (primary 3.862 vs recycled 2.212) 1.65 kgCO2e/kg. Mean 1.41 x 0.02 kg = 0.028. Tiny per item - the honest message is that not buying the bottle saves three to four times more than recycling it. If the alternative is incineration rather than landfill the benefit is larger (burning plastic releases about 1.4 kgCO2e/kg net in EPA WARM). Bottle weight is an assumption.	2023US Environmental Protection Agency 2026UK Department for Energy Security and Net Zero
Recycled paper or cardboard WasteLow	approximately 1.5 kg CO₂e per kg	0.29 to 4 kg	landfill, and new paper from virgin fibre Show workingHide working The paper goes to landfill (where it rots and releases methane) and new paper is made from virgin fibre. UK 2026: recycled-vs-virgin fibre 0.295 + avoided landfill 1.16 = 1.45 kgCO2e/kg (central). Low = fibre benefit only (paper would have been burned, not landfilled). High = EPA WARM v16 mixed paper (3.99), which also credits carbon left standing in forests. The two official methods disagree by a factor of ten because of how they treat landfill methane and forest carbon - hence "low" confidence. One A4 sheet (80 g/m2) weighs 5 g, so recycling a sheet is worth about 7 g CO2e; using less paper saves more.	2026UK Department for Energy Security and Net Zero 2023US Environmental Protection Agency
Composted food scraps WasteMedium	approximately 0.7 kg CO₂e per kg	0 to 0.72 kg	the same scraps rotting in landfill Show workingHide working The scraps go to landfill, where they decompose without air and release methane. EPA WARM v16: landfill 0.5 - composting (-0.15) = 0.65 MTCO2e/short ton = 0.72 kgCO2e/kg. UK 2026: landfill 0.7 - composting 0.009 = 0.69. Mean 0.7. ZERO if your council burns residual waste or already collects food waste for composting or biogas - the benefit exists only relative to landfill (hence low = 0). Home compost must be kept aerated: a soggy, airless heap makes methane too. Not eating-quality food? Then "food waste avoided" does not apply; if it was edible, avoiding the waste saves about five times more.	2023US Environmental Protection Agency 2026UK Department for Energy Security and Net Zero
Brought my own bag WasteMedium	approximately 19 g CO₂e per item	10 to 25 g	a single-use plastic carrier bag Show workingHide working One lightweight (HDPE) supermarket carrier bag, of which about 40% would have been reused once as a bin liner. 1.578 kgCO2e for the 82.14 bags needed for a month's shopping = 0.0192 kgCO2e per bag. Low = every bag reused as a bin liner (0.83); high = none reused (2.082). About 19 g CO2e - the climate benefit is real but tiny; the main win is less plastic litter. The reusable bag must earn its keep first: versus single-use bags a paper bag needs 3 uses, a "bag for life" 4, a woven PP bag 11 and a cotton tote 131 uses to break even on climate. 2006 UK data.	2011Environment Agency (England)
Used my own cup WasteLow	approximately 20 g CO₂e per item	10 to 30 g	a disposable cup Show workingHide working One disposable paper coffee cup; the reusable cup's own manufacture and washing are netted off. UK: 75 ktCO2e a year from about 7 million paper cups a day = 29.4 g CO2e per cup; reusable cups give a threefold reduction, so the net saving is two thirds = 19.6 g. About 20 g CO2e - very small. Derived from the abstract of a single UK study (full text not checked), so low confidence. A reusable cup only wins if it is used many times and washed efficiently. The larger part of a coffee's footprint is the milk.	2020Journal of Cleaner Production
Refilled my bottle WasteLow	approximately 80 g CO₂e per item	40 g to 0.1 kg	a 500 ml bottle of bought water Show workingHide working One new 500 ml PET bottle (the bottle only - its resin and moulding). 0.02 kg bottle x 3.862 kgCO2e/kg (UK 2026, virgin PET including forming) = 0.077. Low = a 9.25 g water bottle; high = top of the life-cycle range compiled by NIH (resin 0.0379-0.0625 + bottle making 0.034-0.046 kgCO2e). Bottle only. Transport and refrigeration of bottled water add more (not quantified here); tap water is about 0.18 g CO2e per 500 ml. Bottle weight is an assumption. EPA WARM gives a lower "source reduction" value for PET (2.39 kgCO2e/kg) because it assumes some recycled content.	2026UK Department for Energy Security and Net Zero 2019US National Institutes of Health 2023US Environmental Protection Agency
Took e-waste or batteries to a collection point WasteNot quantified	Not quantified	No number shown	not quantified: it prevents fires and recovers metals
Planted a tree NatureLow	approximately 6 kg CO₂e per tree year	3.9 to 19 kg	no tree; uptake is slow and only counts while the tree lives Show workingHide working No tree. Average for an urban or garden tree over its first ten years, already discounted for the roughly 40% of seedlings that die. EPA / US DOE: 0.06 tCO2 per urban tree seedling grown for 10 years (36.4 lb carbon x 44/12). We re-derived this from the DOE (1998) table: 60.5 kgCO2 over 10 years = 6.05 kgCO2 per tree-year. Low = conifer (3.9); high = a surviving ten-year-old hardwood (18.6 kgCO2/yr). CAVEATS: (1) Carbon is taken up slowly over decades, not on planting day - credit a tree per year it is alive, never up front. (2) It is storage, not avoided emissions: it is reversed if the tree dies, burns or is felled, so never present it as cancelling out fossil emissions. (3) Young trees absorb little; a surviving moderate-growth hardwood takes up about 38.6 kgCO2/yr at age 20. (4) Figures are for US urban trees, not forests or the tropics. NOTE: the current EPA page labels the 36.4 lb figure "per year"; the DOE table it cites shows this is the 10-year total (our re-calculation is saved in data/raw/doe_1998_tree_check.json).	2026US Environmental Protection Agency 1998US Department of Energy, Energy Information Administration
Picked up litter NatureNot quantified	Not quantified	No number shown	not quantified Show workingHide working Not applicable. No CO2e claim. IMPACT NOT QUANTIFIED - award XP only. Litter picking protects wildlife and waterways and builds community; we found no credible way to express it in CO2e and will not invent one. Show "Impact: not measured in CO2e".
Volunteered for nature NatureNot quantified	Not quantified	No number shown	not quantified Show workingHide working Not applicable. No CO2e claim. IMPACT NOT QUANTIFIED - award XP only. Benefits are for biodiversity, soil, water and people; any carbon effect is site-specific, slow and cannot be attributed to one volunteer-hour.
Tended plants, garden or balcony NatureNot quantified	Not quantified	No number shown	not quantified
Helped wildlife (feeder, bee hotel, a wild corner) NatureNot quantified	Not quantified	No number shown	not quantified
Had a real conversation about climate NatureNot quantified	Not quantified	No number shown	not quantified
Contacted a representative, signed, or showed up NatureNot quantified	Not quantified	No number shown	not quantified

Walked or cycled instead of drivingapproximately 0.21 kg CO₂e per km
MoveHigh
Estimate
approximately 0.21 kg CO₂e
per km
Varies by region
Likely range
0.14 to 0.34 kg
Compared with
the same trip in an average car
Show workingHide working
The same trip driven in an average car (UK fleet average, unknown fuel) with you as the driver. Counts fuel burned plus fuel production ("well-to-wheel"); excludes car manufacture.
km x (tailpipe 0.16591 + fuel production 0.04399) = km x 0.2099 kgCO2e. Low = small hybrid 0.1441; high = large petrol car 0.3413.
Only log trips that would otherwise have been driven. US fleet average is higher: 0.244 kgCO2e/km tailpipe only (EPA, 22.8 mpg). India: a petrol hatchback is about 0.14 and a scooter about 0.0368 kgCO2/km (India GHG Program 2015, tailpipe CO2 only) - use the user's real vehicle. Cross-check: Wynes & Nicholas (2017) put living car-free at 2.4 tCO2e/yr.
Sources
2026UK Department for Energy Security and Net Zero
2026US Environmental Protection Agency
2015India GHG Program
2017Environmental Research Letters
Took the bus instead of drivingapproximately 82 g CO₂e per km
MoveMedium
Estimate
approximately 82 g CO₂e
per km
Varies by region
Likely range
53 g to 0.16 kg
Compared with
driving it alone, minus the bus’s own emissions
Show workingHide working
The same distance driven alone in an average car, compared with the average emissions per passenger-km of a local bus at typical UK occupancy.
km x (car 0.2099 - average local bus 0.128) = km x 0.0819. Low uses a local bus outside London (0.1573); high uses a coach (0.046).
Bus emissions per passenger depend almost entirely on how full the bus is. US average transit bus: 0.041 kgCO2/passenger-km (EPA 2025); Indian city bus: 0.015161 kgCO2/passenger-km (India GHG Program 2015). If the alternative would have been a motorbike or a shared car the saving is smaller and can be zero.
Sources
2026UK Department for Energy Security and Net Zero
2025US Environmental Protection Agency
2015India GHG Program
Took the train or metro instead of drivingapproximately 0.17 kg CO₂e per km
MoveHigh
Estimate
approximately 0.17 kg CO₂e
per km
Varies by region
Likely range
0.13 to 0.2 kg
Compared with
driving it alone, minus the train’s own emissions
Show workingHide working
The same distance driven alone in an average car, compared with average national rail emissions per passenger-km.
km x (car 0.2099 - national rail 0.0399) = km x 0.17. Low uses US commuter rail (0.0826 kgCO2/pkm, EPA 2025); high uses electric international rail (0.0125).
Metro and tram are similar or better: London Underground 0.0228, light rail/tram 0.0287 kgCO2e/passenger-km (UK 2026). Indian Railways: about 0.007837 kgCO2/passenger-km (2014-15).
Sources
2026UK Department for Energy Security and Net Zero
2025US Environmental Protection Agency
2015India GHG Program
Shared the rideapproximately 0.21 kg CO₂e per km
MoveMedium
Estimate
approximately 0.21 kg CO₂e
per km
Varies by region
Likely range
40 g to 0.34 kg
Compared with
everyone driving alone; your share of the car-km avoided
Show workingHide working
Each extra passenger would otherwise have driven the same distance alone in an average car. One car-kilometre is avoided per shared kilometre per extra passenger.
km x extra passengers x 0.2099 (one average car-km avoided). Low: the passenger would otherwise have taken the train (0.0399); high: a large petrol car is left at home (0.3413).
DOUBLE-COUNTING RISK: the saving belongs to the car, not to each person. Log it once (the driver logs the passengers, or each rider logs a 1/n share). Detours to pick people up reduce the saving and are not modelled.
Sources
2026UK Department for Energy Security and Net Zero
Rode my e-bike or e-scooter instead of drivingapproximately 0.21 kg CO₂e per km
MoveHigh
Estimate
approximately 0.21 kg CO₂e
per km
Varies by region
Likely range
0.13 to 0.34 kg
Compared with
the same trip in an average car, minus charging
Show workingHide working
The same trip driven alone in an average car. The e-bike uses a little grid electricity (about 0.01 kWh per km).
km x (car 0.2099 - 0.01 kWh x grid). At the world-average grid (0.45849 kg/kWh) that is 0.2053 per km. kWhPerUnit is negative because the action USES electricity.
E-bike electricity is tiny: McQueen et al. (2019) calculate 3.778-12.568 gCO2 per mile across US grid regions (4.905 in the US Northwest); a 160 Wh pedelec battery giving 25-35 km implies 0.005-0.006 kWh/km (Petrillo et al. 2018). 0.01 kWh/km is our rounded, deliberately cautious assumption. Applies to PRIVATELY OWNED e-bikes and e-scooters. Shared dockless e-scooters are different: Hollingsworth et al. (2019) found 202 gCO2e per passenger-mile over the life cycle (vs 414 for a car) and only 34% of rides replaced a car trip - do not award this factor for shared scooters that replaced walking.
Sources
2026UK Department for Energy Security and Net Zero
2019Portland State University, Transportation Research and Education Center
2018IntechOpen
2019Environmental Research Letters
2026Ember
Drove electric instead of petrolapproximately 0.11 kg CO₂e per km
MoveMedium
Estimate
approximately 0.11 kg CO₂e
per km
Varies by region
Likely range
71 g to 0.2 kg
Compared with
the same distance in an average petrol car, minus charging
Show workingHide working
The same distance in an average petrol car. The electric car draws 0.20358 kWh per km from the grid (UK average battery-electric car).
km x (petrol car 0.2075 - 0.20358 kWh x grid). World-average grid: 0.114; India: 0.071; France: 0.199.
Operating emissions only: battery and vehicle manufacturing are excluded on both sides (an EV carries a larger manufacturing footprint that is repaid over the first years of driving). Grid losses (5-10%) are not added. The UK Government's own all-scope figure for an average battery-electric car on the 2026 UK grid is 0.04 kgCO2e/km.
Sources
2026UK Department for Energy Security and Net Zero
2026Ember
Worked from home instead of commuting by carapproximately 3.5 kg CO₂e per day
MoveLow
Estimate
approximately 3.5 kg CO₂e
per day
Varies by region
Likely range
0 to 12 kg
Compared with
your car commute, minus the extra energy used at home
Show workingHide working
A return car commute of 29.2 km (England average commuting trip of 14.6 km each way) minus the extra heating and equipment energy used at home over an 8-hour day.
2 x 14.6 km x 0.2099 - 8 h x 0.32393 kgCO2e/h (UK homeworking factor: heating 0.30234 + equipment 0.02159) = 3.54. Low = same distance by train (-1.43, i.e. working from home is WORSE). High = double the distance by car with no extra heating (12.09).
Can be NEGATIVE. The app should ask for the real round-trip distance and mode and compute: roundTripKm x modeFactor - hours x 0.324 (heating season) or x 0.022 (no heating or cooling needed). Break-even for a car commuter is about 6.2 km each way, which matches the IEA finding that home-working only cuts CO2 for car commutes longer than about 6 km (IEA text seen only as a search excerpt - site blocks automated access). The UK heating factor is an annual average for gas-heated homes; air-conditioning in hot climates is not covered.
Sources
2026UK Department for Energy Security and Net Zero
2026UK Department for Energy Security and Net Zero
2025UK Department for Transport
2020International Energy Agency
Took the train instead of a short flightapproximately 0.11 kg CO₂e per km
MoveMedium
Estimate
approximately 0.11 kg CO₂e
per km
Likely range
57 g to 0.22 kg
Compared with
an economy seat on a short-haul flight, minus the train
Show workingHide working
An economy seat on a short-haul flight over the same great-circle distance, including the non-CO2 warming effects of flying at altitude, compared with national rail.
km x (short-haul economy flight 0.1483 - national rail 0.0399) = km x 0.1084. Low = flight without non-CO2 effects (0.0968); high = UK domestic flight average (0.2628).
UK flight factors include an 8% uplift for indirect routing and a 1.7 multiplier on CO2 for non-CO2 effects ("radiative forcing"); the multiplier is scientifically uncertain. Use the great-circle distance, not the (longer) rail distance. Electric high-speed rail is cleaner than the national-rail average used here (0.0125 for international rail).
Sources
2026UK Department for Energy Security and Net Zero
2026UK Department for Energy Security and Net Zero
Took the train instead of a short flight (distance unknown)approximately 54 kg CO₂e per trip
MoveMedium
Estimate
approximately 54 kg CO₂e
per trip
Likely range
28 to 110 kg
Compared with
a one-way short-haul flight of about 500 km, minus the train
Show workingHide working
One economy seat on a 500 km one-way flight (including non-CO2 effects) compared with the same journey by national rail.
500 km x 0.1084 = 54.2 kgCO2e one way. For a specific route use the per-km action with the real great-circle distance.
Illustrative distance. Approximate great-circle distances (our haversine arithmetic): London-Paris 347 km, London-Edinburgh 534 km, San Francisco-Los Angeles 544 km, Sydney-Melbourne 706 km, Delhi-Mumbai 1,137 km. A return trip is two units.
Sources
2026UK Department for Energy Security and Net Zero
2026UK Department for Energy Security and Net Zero
Car-free dayNot quantified
MoveNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified: the kilograms come from the trips you log
Sources
Plant-based mealapproximately 1.5 kg CO₂e per meal
EatMedium
Estimate
approximately 1.5 kg CO₂e
per meal
Likely range
0.97 to 2.6 kg
Compared with
a typical meal with meat
Show workingHide working
One of three daily meals of a medium meat-eater (50-99 g of meat a day), replaced by a fully plant-based (vegan) meal.
(medium meat-eater 7.04 - vegan 2.47 kgCO2e/day) / 3 meals = 1.52. Low uses a low meat-eater (5.37), high a high meat-eater (10.24).
Scarborough et al. linked the real diets of 55,504 UK adults to the Poore & Nemecek life-cycle database (farm to retail, global-average production, land-use change included, GWP100), standardised to 2,000 kcal/day. Assuming three equal meals is our simplification. Do not also log a "vegetarian day" for the same day.
Sources
2023Nature Food
2018Our World in Data / Science
Plant-based instead of a beef mealapproximately 6.3 kg CO₂e per meal
EatMedium
Estimate
approximately 6.3 kg CO₂e
per meal
Likely range
3 to 9.8 kg
Compared with
the same meal made with about 100 g of beef
Show workingHide working
A portion of beef supplying 20 g of protein (about 100 g of meat), replaced by tofu supplying the same protein.
Per 20 g protein (Poore & Nemecek global means): beef from beef herds 9.98, beef from dairy herds 3.37, tofu 0.4, pulses 0.17 kgCO2e. Central = mean of the two beef types minus tofu = 6.28. Low = dairy-herd beef vs tofu; high = beef-herd beef vs pulses.
Beef is the single largest lever in most diets, but its footprint varies enormously by farm and country: the global means are 99.48 (beef herd) and 33.3 (dairy herd) kgCO2e per kg, including land-use change and methane. The central value is an unweighted mean of those two - we could not verify the production split, so treat it as indicative. Many national beef footprints (without land-use change) are lower.
Sources
2018Our World in Data / Science
2018Our World in Data / Science
2018Our World in Data / Science
Chicken instead of beefapproximately 5.7 kg CO₂e per serving
EatMedium
Estimate
approximately 5.7 kg CO₂e
per serving
Likely range
2.3 to 9 kg
Compared with
the same 100 g serving as beef
Show workingHide working
100 g of beef (retail weight) replaced by 100 g of poultry meat.
0.1 kg x (beef 66.39 [mean of 99.48 and 33.3] - poultry 9.87) = 5.65 kgCO2e.
Shows that the choice of meat matters almost as much as going meat-free. Lamb (39.72) and cheese (23.88) are also high; pork is 12.31 kgCO2e/kg (global means).
Sources
2018Our World in Data / Science
Vegetarian dayapproximately 2.9 kg CO₂e per day
EatMedium
Estimate
approximately 2.9 kg CO₂e
per day
Likely range
1.2 to 6.1 kg
Compared with
a day of typical meals with meat
Show workingHide working
A normal day of eating for a medium meat-eater (50-99 g of meat a day), compared with a typical vegetarian day.
medium meat-eater 7.04 - vegetarian 4.16 = 2.88 kgCO2e/day. Low: low meat-eater (5.37); high: high meat-eater (10.24).
Vegetarian diets that lean heavily on cheese and other dairy save less. Do not also log individual plant-based meals for the same day.
Sources
2023Nature Food
Fully plant-based dayapproximately 4.6 kg CO₂e per day
EatMedium
Estimate
approximately 4.6 kg CO₂e
per day
Likely range
2.9 to 7.8 kg
Compared with
a day of typical meals with meat
Show workingHide working
A normal day of eating for a medium meat-eater, compared with a typical vegan day.
medium meat-eater 7.04 - vegan 2.47 = 4.57 kgCO2e/day.
Over a year this is 1.67 tCO2e. Wynes & Nicholas (2017) estimate 0.8 tCO2e/yr for a plant-based diet using older, lower food footprints - the two sources bracket the plausible range.
Sources
2023Nature Food
2017Environmental Research Letters
Plant milk instead of dairyapproximately 2.2 kg CO₂e per litre
EatHigh
Estimate
approximately 2.2 kg CO₂e
per litre
Likely range
2 to 2.4 kg
Compared with
the same amount of cow’s milk
Show workingHide working
One litre of dairy milk replaced by one litre of plant milk (average of soy, oat, almond and rice).
dairy 3.15 - mean plant milk 0.942 (soy 0.98, oat 0.9, almond 0.7, rice 1.18) = 2.21 kgCO2e per litre.
A 200 ml glass is one fifth of this. Almond milk has the lowest carbon footprint of the four but by far the highest water use; oat and soy are good all-rounders.
Sources
2018Our World in Data / Science
Saved food from the binapproximately 3.6 kg CO₂e per kg
EatMedium
Estimate
approximately 3.6 kg CO₂e
per kg
Likely range
0.84 to 4.6 kg
Compared with
binning it and buying the same food again
Show workingHide working
The food would have been thrown away and replaced by buying the same amount again. Counts the whole supply chain of the wasted food.
UK household food waste: 16 MtCO2e / 4.4 Mt edible food = 3.64 kgCO2e per kg (WRAP 2022 data). Cross-checks: UK Government 2026 "food and drink" factor 3.7; EPA WARM v16 source reduction 4.03. Low = EPA non-meat food (0.84); high = EPA mixed food plus avoided landfill methane (4.59).
What was saved matters far more than how much: EPA WARM gives 0.49 kgCO2e/kg for fruit and vegetables, 0.73 for bread, 1.93 for dairy and 33.2 for beef. Do not add "composting" for the same food.
Sources
2025WRAP
2026UK Department for Energy Security and Net Zero
2023US Environmental Protection Agency
Rescued a meal (leftovers, food-rescue app)approximately 1.5 kg CO₂e per meal
EatMedium
Estimate
approximately 1.5 kg CO₂e
per meal
Likely range
0.35 to 1.9 kg
Compared with
binning the meal and cooking another
Show workingHide working
A 420 g meal that would otherwise have been binned and replaced.
0.42 kg (WRAP standard meal) x 3.64 kgCO2e/kg = 1.53 kgCO2e.
Same evidence as "food waste avoided" expressed per meal - log one or the other, never both. Commercial food-rescue apps often quote 2.5-2.7 kgCO2e per meal; that is at the top of our range.
Sources
2020WRAP
2025WRAP
2023US Environmental Protection Agency
Chose local, seasonal produceapproximately 70 g CO₂e per meal
EatLow
Estimate
approximately 70 g CO₂e
per meal
Likely range
0 to 0.1 kg
Compared with
the same meal with ingredients shipped the usual distances
Show workingHide working
The same meal made from ingredients transported the usual distances by road and sea.
Transport is about 6% of diet emissions (EU diets, Sandstrom et al. 2018 via Our World in Data). 6% x 2.35 kgCO2e (a medium meat-eater's average meal) = 0.14 if ALL transport vanished (high). Central assumes half of it is avoided: 0.07.
HONEST VERSION: the climate effect is small - what you eat matters far more than where it came from. Transport is 0.5% of beef's footprint and under 10% for most foods. Local food grown in heated greenhouses or stored for months can be worse than imports. The real exception is AIR-FREIGHTED produce (out-of-season asparagus, green beans, berries): air freight emits 1.13 kgCO2e per tonne-km against 0.023 by sea, so 1 kg flown 8,000 km adds about 9 kgCO2e - but only 0.16% of food miles are flown. Reward this action for its other benefits (freshness, local growers) and show the number as "small".
Sources
2020Our World in Data
2020Our World in Data
2018Our World in Data / Science
2023Nature Food
Heating 1 °C lowerapproximately 0.77 kg CO₂e per day
PowerLow
Estimate
approximately 0.77 kg CO₂e
per day
Varies by region
Likely range
0.5 to 2.2 kg
Compared with
the same home kept 1 °C warmer today
Show workingHide working
The same gas-heated home kept 1 degC warmer on a heating-season day.
Typical UK home: 11500 kWh gas/yr (Ofgem median) x 82% space heating = 9430 kWh; x 7.000000000000001% per degC (IEA/European Commission) = 660 kWh/yr; / 182 heating days = 3.63 kWh gas per day; x 0.21252 kgCO2e/kWh = 0.77. Low = Ofgem low-use home (7500 kWh). High = Energy Saving Trust's 22->21 degC saving of GBP 120/yr at 6.5p/kWh = 1846 kWh/yr.
PER HOUSEHOLD, not per person - one household member logs it, or divide by the number of people. Only valid on days the heating is actually on. Very climate- and building-dependent (the two sources we use differ by a factor of three). heatKWhPerUnit here is kWh of GAS, not useful heat. For electric heating multiply the kWh by your grid intensity instead (and divide by about 3 for a heat pump). Never suggest going below a healthy indoor temperature (WHO recommends at least 18 degC).
Sources
2022International Energy Agency / European Commission
2023Ofgem
2026Energy Saving Trust
2026Energy Saving Trust
2026Energy Saving Trust
2026UK Department for Energy Security and Net Zero
2026UK Department for Energy Security and Net Zero
Air-dried a loadapproximately 0.69 kg CO₂e per load
PowerMedium
Estimate
approximately 0.69 kg CO₂e
per load
Varies by region
Likely range
41 g to 2 kg
Compared with
the same load in a tumble dryer
Show workingHide working
The same load dried in an electric tumble dryer (average measured UK cycle: 1.52 kWh).
394 kWh/yr / 260 cycles = 1.515 kWh per measured drying cycle; x world-average grid 0.45849 = 0.69 kgCO2e.
Measured UK cycles vary widely: 33% used under 1 kWh, 75% under 2 kWh and 10% over 3 kWh (many are short top-up cycles). A full load in a standard US electric dryer that just meets ENERGY STAR uses about 2.15 kWh (8.45 lb test load / 3.93 lb per kWh); heat-pump dryers use roughly half. Low = 1 kWh on the French grid; high = 3 kWh on the Indian grid. Only counts if you own and would have used a dryer. Wynes & Nicholas (2017): 0.21 tCO2e/yr.
Sources
2012UK Government (DEFRA / DECC / Energy Saving Trust)
2014US EPA ENERGY STAR
2026Energy Saving Trust
2026Ember
2017Environmental Research Letters
Switched standby devices off at the wallapproximately 0.23 kg CO₂e per day
PowerLow
Estimate
approximately 0.23 kg CO₂e
per day
Varies by region
Likely range
10 g to 0.76 kg
Compared with
the household’s devices left on standby all day
Show workingHide working
All the household's TVs, consoles, set-top boxes, chargers and similar devices left on standby for the day.
Energy Saving Trust: about GBP 45 a year saved at 24.7p/kWh = 182 kWh/yr = 0.5 kWh per day (an average draw of 21 W); x world-average grid = 0.23 kgCO2e.
PER HOUSEHOLD and derived from a money figure, so treat as rough. High = the 47 W average standby load measured in English homes in 2010-11 (1.13 kWh/day) on the Indian grid; today's devices draw far less (EU rules cap standby at 0.5-1 W per device). Low = half the Energy Saving Trust figure on the French grid. Unplugging a single phone charger saves almost nothing.
Sources
2026Energy Saving Trust
2026Energy Saving Trust
2012UK Government (DEFRA / DECC / Energy Saving Trust)
2026Ember
Replaced an old bulb with an LEDapproximately 33 g CO₂e per bulb day
PowerMedium
Estimate
approximately 33 g CO₂e
per bulb day
Varies by region
Likely range
1 to 91 g
Compared with
a 60 W bulb left in the fitting (day one only is counted)
Show workingHide working
A 60 W incandescent (or equivalent halogen) bulb left in the same fitting, used the US-average 1.6 hours a day.
60 W x 75% saving (LEDs use at least 75% less than incandescent, US DOE) x 1.6 h/day (US DOE/PNNL average per lamp) = 0.072 kWh per bulb per day; x world-average grid = 0.033 kgCO2e. That is 26.3 kWh a year per bulb.
A one-off action with a small ongoing benefit: award XP once, then count the saving passively. If the old bulb was already a compact fluorescent the saving is close to zero. Low = 40 W bulb used 1 h/day on the French grid; high = 60 W bulb used 3 h/day (typical for outdoor lights) on the Indian grid. Energy Saving Trust: LEDs use 80% less electricity than halogens. The "at least 75% less" DOE statement was only seen as a search excerpt (page moved).
Sources
2012US Department of Energy / Pacific Northwest National Laboratory
2024US Department of Energy
2026Energy Saving Trust
2026Ember
AC 1 °C warmer, or a fan insteadNot quantified
PowerNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified: we have no sourced factor for cooling yet
Sources
Shorter showerapproximately 52 g CO₂e per minute
WaterMedium
Estimate
approximately 52 g CO₂e
per minute
Varies by region
Likely range
31 g to 0.15 kg
Compared with
the same shower, that many minutes longer
Show workingHide working
One more minute under a hot shower flowing at 7.6 litres a minute, heated by a gas boiler.
7.57 L/min (2 US gal/min) x 4.186 kJ/kg.K x 25 K / 3600 = 0.22 kWh of heat per minute; / 0.9 boiler efficiency x 0.21252 kgCO2e per kWh gas = 0.052 kgCO2e. Electric water heater: 0.22 kWh x grid (world average: 0.101).
Physics-based: flow rate x temperature rise. The 25 K rise (e.g. 15 -> 40 degC) and 90% boiler efficiency are our assumptions. Low = 1.5 gal/min low-flow head, 20 K rise, gas; high = standard 2.5 gal/min head, 30 K rise, electric resistance heater on the world-average grid. heatKWhPerUnit is USEFUL HEAT: divide by 0.9 and multiply by the gas factor, or multiply by the grid intensity for electric heaters (divide by about 3 for a heat pump). Solar water heaters and unheated showers: zero.
Sources
2026US Environmental Protection Agency
2023US National Institute of Standards and Technology
2026UK Department for Energy Security and Net Zero
2026Energy Saving Trust
2026Ember
Saved hot waterapproximately 7 g CO₂e per litre
WaterMedium
Estimate
approximately 7 g CO₂e
per litre
Varies by region
Likely range
0 to 19 g
Compared with
heating that many more litres
Show workingHide working
One more litre of water heated by 25 degC in a gas boiler, then supplied and treated as waste water.
4.186 kJ/kg.K x 25 K / 3600 = 0.0291 kWh of heat per litre; / 0.9 x 0.21252 = 0.0069 kgCO2e, plus 0.00036 for water supply and treatment (UK 2026) = 0.0072.
ENERGY BASIS: almost all of the footprint of water use at home is the energy to HEAT it - about 19 times more than pumping and treating it. Use for baths (a 80-litre bath is 80 units), washing-up and hot taps. Low = heat pump on the French grid; high = electric resistance heater on the Indian grid. Do not also log "shorter shower" for the same water.
Sources
2023US National Institute of Standards and Technology
2026UK Department for Energy Security and Net Zero
2026Ember
Washed at 30 °C instead of 40 °Capproximately 0.1 kg CO₂e per load
WaterMedium
Estimate
approximately 0.1 kg CO₂e
per load
Varies by region
Likely range
5 g to 0.34 kg
Compared with
the same load at 40 °C
Show workingHide working
The same load washed on a 40 degC cycle (average measured UK cycle: 0.585 kWh).
Average measured wash cycle 166 kWh/yr / 284 cycles = 0.585 kWh; washing at 30 degC uses 38% less (Which? lab tests) = 0.222 kWh saved per load. kgCO2e = kWh x grid intensity (world average 0.45849: 0.102). Dropping from 60 degC to 30 degC saves at least 0.51 kWh.
kWh is the robust part; kgCO2e depends on the grid. Low = an efficient machine saving half as much on a very clean grid (France); high = 60->30 degC on a coal-heavy grid (India). The measured cycle average is from 2010-11 UK machines; modern machines use somewhat less. Wynes & Nicholas (2017) put cold-water washing at 0.25 tCO2e/yr for North American households.
Sources
2012UK Government (DEFRA / DECC / Energy Saving Trust)
2026Which?
2026Ember
2017Environmental Research Letters
Washed cold instead of 40 °Capproximately 0.17 kg CO₂e per load
WaterMedium
Estimate
approximately 0.17 kg CO₂e
per load
Varies by region
Likely range
8 g to 0.44 kg
Compared with
the same load at 40 °C
Show workingHide working
The same load washed on a 40 degC cycle.
0.585 kWh x 62% (Which? lab tests) = 0.362 kWh saved per load; x world-average grid 0.45849 = 0.166 kgCO2e.
Log either this or the 30 degC action for a load, not both. Health guidance still recommends an occasional hot wash for towels, bedding and when someone is ill.
Sources
2012UK Government (DEFRA / DECC / Energy Saving Trust)
2026Which?
2026Ember
Tap off while brushingapproximately 10 g CO₂e per day
WaterMedium
Estimate
approximately 10 g CO₂e
per day
Likely range
4 to 10 g
Compared with
the tap left running, morning and evening
Show workingHide working
The cold tap left running while brushing, morning and evening.
8 US gallons a day (EPA WaterSense) = 30.3 litres x 0.362 g CO2e per litre (UK 2026 water supply + treatment) = 11 g CO2e.
HONEST VERSION: about 11 g CO2e a day - negligible for the climate. It saves roughly 30 litres of drinking water, which matters where water is scarce. Show litres saved, not CO2e. High = 9 L/min tap (UK Consumer Council for Water via Energy Saving Trust) for 2 x 2 minutes; the water-supply factor is UK-specific.
Sources
2026US Environmental Protection Agency
2026UK Department for Energy Security and Net Zero
2026Energy Saving Trust
Bought a top second-handapproximately 2.9 kg CO₂e per item
StuffLow
Estimate
approximately 2.9 kg CO₂e
per item
Likely range
2.2 to 7.5 kg
Compared with
the same garment bought new
Show workingHide working
A new T-shirt would have been made and bought. WRAP finds that 65% of second-hand purchases really displace a new one.
0.2 kg x 22.31 kgCO2e per kg of new clothing (UK 2026) = 4.46; x 0.646 displacement = 2.88 kgCO2e. Low = 150 g shirt; high = WRAP's "over 7.5 kg" for a cotton T-shirt.
Garment weight (200 g) is an assumption. If the purchase is an extra you would not otherwise have bought, the saving is zero - that is what the displacement rate captures on average. Shipping of second-hand items is not deducted.
Sources
2026UK Department for Energy Security and Net Zero
2025WRAP
Bought jeans second-handapproximately 13 kg CO₂e per item
StuffMedium
Estimate
approximately 13 kg CO₂e
per item
Likely range
10 to 30 kg
Compared with
the same jeans bought new
Show workingHide working
A new pair of jeans would have been made and bought; 65% of second-hand purchases displace a new one (WRAP).
Levi's 501 life-cycle assessment: fibre 2.9 + fabric 9 + cut/sew/finish 2.6 + sundries/packaging 1.7 + transport/retail 3.8 = 20 kgCO2e before the customer wears them (of 33.4 over the whole life); x 0.646 = 12.9. High = WRAP's "over 30 kg" claim.
Washing and drying the jeans (12.5 kgCO2e over their life in the Levi's study) happens either way and is not counted.
Sources
2015Levi Strauss & Co.
2025WRAP
2026UK Department for Energy Security and Net Zero
Repaired instead of replacingapproximately 7 kg CO₂e per item
StuffLow
Estimate
approximately 7 kg CO₂e
per item
Likely range
3.7 to 73 kg
Compared with
buying a new replacement
Show workingHide working
A new replacement would have been bought. WRAP finds 82% of repairs genuinely prevent a new purchase.
Default (small electrical item, 1.5 kg): 1.5 kg x 5.648 kgCO2e/kg (UK 2026) = 8.5; x 0.822 displacement = 7 kgCO2e. See variants for other item classes.
Item-class dependent - use the variants. The displacement rate was measured for clothing and is applied to other items by analogy (assumption). Item weights are assumptions. The repair itself (parts, travel) is not deducted.
Sources
2025WRAP
2026UK Department for Energy Security and Net Zero
2015Levi Strauss & Co.
2025Apple
2025Apple
Borrowed or rented instead of buyingapproximately 7.6 kg CO₂e per item
StuffLow
Estimate
approximately 7.6 kg CO₂e
per item
Likely range
0.36 to 10 kg
Compared with
buying the item new
Show workingHide working
Buying the item new for a one-off use. The shared item is assumed to serve about ten borrowers.
Default (tool or small appliance, 1.5 kg): 8.5 kgCO2e to make (UK 2026 small electricals 5.648 kgCO2e/kg) x 0.9 = 7.6. Low = a 300 g library book (paper 1.344 kgCO2e/kg); high = a 500 g outfit (clothing 22.31 kgCO2e/kg).
ASSUMPTION-DRIVEN: item weights and the "ten borrowers" share are ours, there is no measured displacement rate for lending, and rental delivery or dry-cleaning is not deducted. Only log when you genuinely would have bought the item. Use for XP and storytelling more than for the CO2e total.
Sources
2026UK Department for Energy Security and Net Zero
Keeping my phone another yearapproximately 15 kg CO₂e per year
StuffMedium
Estimate
approximately 15 kg CO₂e
per year
Likely range
11 to 22 kg
Compared with
replacing it now with a new phone
Show workingHide working
Replacing the phone now with a new one that would itself be used for three years.
iPhone 17 (256 GB): 55 kgCO2e life cycle, of which production 76% + transport 4% = 44 kgCO2e before first use. Delaying replacement by a year avoids one year's share: 44 / 3 = 14.7. Low = 4-year replacement cycle; high = 2-year cycle.
A once-a-year action, not a daily one - award it once per year. One manufacturer's own figure for one model; other phones range from roughly 40 to 90 kgCO2e. Charging a phone for a year is only a few kgCO2e, so keeping the old one is almost pure gain. A laptop is worth about 22 kgCO2e per extra year (MacBook Air M4: 120 kgCO2e, 74% before first use, 4-year life).
Sources
2025Apple
Sold, donated or swapped somethingNot quantified
StuffNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified: the saving belongs to whoever uses it next
Sources
Recycled aluminium cansapproximately 0.13 kg CO₂e per item
WasteMedium
Estimate
approximately 0.13 kg CO₂e
per item
Likely range
81 g to 0.18 kg
Compared with
landfill, and new cans made from ore
Show workingHide working
The can goes to landfill and new cans are made from virgin aluminium instead of recycled metal.
Per kg: EPA WARM v16 (recycling -9.13 vs landfill 0.02 MTCO2e/short ton) = 10.09 kgCO2e/kg; UK 2026 (primary 9.114 vs recycled 0.994 kgCO2e/kg) = 8.12. Mean 9.11 x 0.014 kg per can = 0.127.
Aluminium is the star of recycling - recycled metal needs a small fraction of the energy of smelting new. The per-kg factor is well evidenced; the 14 g can weight is an UNVERIFIED assumption (range used: 10-17.5 g). Only counts where cans are actually collected and recycled.
Sources
2023US Environmental Protection Agency
2026UK Department for Energy Security and Net Zero
Recycled glass bottles or jarsapproximately 91 g CO₂e per item
WasteMedium
Estimate
approximately 91 g CO₂e
per item
Likely range
66 g to 0.29 kg
Compared with
landfill, and new glass from raw materials
Show workingHide working
The bottle goes to landfill and new glass is made from virgin raw materials.
Per kg: EPA WARM v16 0.33; UK 2026 (primary 1.403 vs recycled 0.823) 0.58 kgCO2e/kg. Mean 0.46 x 0.2 kg = 0.091. High = a 500 g wine bottle at the UK factor.
Small per item because glass is heavy but not carbon-intensive per kg. Refilling or reusing a bottle beats recycling it. The 200 g bottle weight is an assumption.
Sources
2023US Environmental Protection Agency
2026UK Department for Energy Security and Net Zero
Recycled plastic bottlesapproximately 28 g CO₂e per item
WasteMedium
Estimate
approximately 28 g CO₂e
per item
Likely range
11 to 41 g
Compared with
landfill, and new plastic from oil
Show workingHide working
The bottle goes to landfill and new bottles are made from virgin PET.
Per kg PET: EPA WARM v16 1.17; UK 2026 (primary 3.862 vs recycled 2.212) 1.65 kgCO2e/kg. Mean 1.41 x 0.02 kg = 0.028.
Tiny per item - the honest message is that not buying the bottle saves three to four times more than recycling it. If the alternative is incineration rather than landfill the benefit is larger (burning plastic releases about 1.4 kgCO2e/kg net in EPA WARM). Bottle weight is an assumption.
Sources
2023US Environmental Protection Agency
2026UK Department for Energy Security and Net Zero
Recycled paper or cardboardapproximately 1.5 kg CO₂e per kg
WasteLow
Estimate
approximately 1.5 kg CO₂e
per kg
Likely range
0.29 to 4 kg
Compared with
landfill, and new paper from virgin fibre
Show workingHide working
The paper goes to landfill (where it rots and releases methane) and new paper is made from virgin fibre.
UK 2026: recycled-vs-virgin fibre 0.295 + avoided landfill 1.16 = 1.45 kgCO2e/kg (central). Low = fibre benefit only (paper would have been burned, not landfilled). High = EPA WARM v16 mixed paper (3.99), which also credits carbon left standing in forests.
The two official methods disagree by a factor of ten because of how they treat landfill methane and forest carbon - hence "low" confidence. One A4 sheet (80 g/m2) weighs 5 g, so recycling a sheet is worth about 7 g CO2e; using less paper saves more.
Sources
2026UK Department for Energy Security and Net Zero
2023US Environmental Protection Agency
Composted food scrapsapproximately 0.7 kg CO₂e per kg
WasteMedium
Estimate
approximately 0.7 kg CO₂e
per kg
Likely range
0 to 0.72 kg
Compared with
the same scraps rotting in landfill
Show workingHide working
The scraps go to landfill, where they decompose without air and release methane.
EPA WARM v16: landfill 0.5 - composting (-0.15) = 0.65 MTCO2e/short ton = 0.72 kgCO2e/kg. UK 2026: landfill 0.7 - composting 0.009 = 0.69. Mean 0.7.
ZERO if your council burns residual waste or already collects food waste for composting or biogas - the benefit exists only relative to landfill (hence low = 0). Home compost must be kept aerated: a soggy, airless heap makes methane too. Not eating-quality food? Then "food waste avoided" does not apply; if it was edible, avoiding the waste saves about five times more.
Sources
2023US Environmental Protection Agency
2026UK Department for Energy Security and Net Zero
Brought my own bagapproximately 19 g CO₂e per item
WasteMedium
Estimate
approximately 19 g CO₂e
per item
Likely range
10 to 25 g
Compared with
a single-use plastic carrier bag
Show workingHide working
One lightweight (HDPE) supermarket carrier bag, of which about 40% would have been reused once as a bin liner.
1.578 kgCO2e for the 82.14 bags needed for a month's shopping = 0.0192 kgCO2e per bag. Low = every bag reused as a bin liner (0.83); high = none reused (2.082).
About 19 g CO2e - the climate benefit is real but tiny; the main win is less plastic litter. The reusable bag must earn its keep first: versus single-use bags a paper bag needs 3 uses, a "bag for life" 4, a woven PP bag 11 and a cotton tote 131 uses to break even on climate. 2006 UK data.
Sources
2011Environment Agency (England)
Used my own cupapproximately 20 g CO₂e per item
WasteLow
Estimate
approximately 20 g CO₂e
per item
Likely range
10 to 30 g
Compared with
a disposable cup
Show workingHide working
One disposable paper coffee cup; the reusable cup's own manufacture and washing are netted off.
UK: 75 ktCO2e a year from about 7 million paper cups a day = 29.4 g CO2e per cup; reusable cups give a threefold reduction, so the net saving is two thirds = 19.6 g.
About 20 g CO2e - very small. Derived from the abstract of a single UK study (full text not checked), so low confidence. A reusable cup only wins if it is used many times and washed efficiently. The larger part of a coffee's footprint is the milk.
Sources
2020Journal of Cleaner Production
Refilled my bottleapproximately 80 g CO₂e per item
WasteLow
Estimate
approximately 80 g CO₂e
per item
Likely range
40 g to 0.1 kg
Compared with
a 500 ml bottle of bought water
Show workingHide working
One new 500 ml PET bottle (the bottle only - its resin and moulding).
0.02 kg bottle x 3.862 kgCO2e/kg (UK 2026, virgin PET including forming) = 0.077. Low = a 9.25 g water bottle; high = top of the life-cycle range compiled by NIH (resin 0.0379-0.0625 + bottle making 0.034-0.046 kgCO2e).
Bottle only. Transport and refrigeration of bottled water add more (not quantified here); tap water is about 0.18 g CO2e per 500 ml. Bottle weight is an assumption. EPA WARM gives a lower "source reduction" value for PET (2.39 kgCO2e/kg) because it assumes some recycled content.
Sources
2026UK Department for Energy Security and Net Zero
2019US National Institutes of Health
2023US Environmental Protection Agency
Took e-waste or batteries to a collection pointNot quantified
WasteNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified: it prevents fires and recovers metals
Sources
Planted a treeapproximately 6 kg CO₂e per tree year
NatureLow
Estimate
approximately 6 kg CO₂e
per tree year
Likely range
3.9 to 19 kg
Compared with
no tree; uptake is slow and only counts while the tree lives
Show workingHide working
No tree. Average for an urban or garden tree over its first ten years, already discounted for the roughly 40% of seedlings that die.
EPA / US DOE: 0.06 tCO2 per urban tree seedling grown for 10 years (36.4 lb carbon x 44/12). We re-derived this from the DOE (1998) table: 60.5 kgCO2 over 10 years = 6.05 kgCO2 per tree-year. Low = conifer (3.9); high = a surviving ten-year-old hardwood (18.6 kgCO2/yr).
CAVEATS: (1) Carbon is taken up slowly over decades, not on planting day - credit a tree per year it is alive, never up front. (2) It is storage, not avoided emissions: it is reversed if the tree dies, burns or is felled, so never present it as cancelling out fossil emissions. (3) Young trees absorb little; a surviving moderate-growth hardwood takes up about 38.6 kgCO2/yr at age 20. (4) Figures are for US urban trees, not forests or the tropics. NOTE: the current EPA page labels the 36.4 lb figure "per year"; the DOE table it cites shows this is the 10-year total (our re-calculation is saved in data/raw/doe_1998_tree_check.json).
Sources
2026US Environmental Protection Agency
1998US Department of Energy, Energy Information Administration
Picked up litterNot quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Show workingHide working
Not applicable.
No CO2e claim.
IMPACT NOT QUANTIFIED - award XP only. Litter picking protects wildlife and waterways and builds community; we found no credible way to express it in CO2e and will not invent one. Show "Impact: not measured in CO2e".
Sources
Volunteered for natureNot quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Show workingHide working
Not applicable.
No CO2e claim.
IMPACT NOT QUANTIFIED - award XP only. Benefits are for biodiversity, soil, water and people; any carbon effect is site-specific, slow and cannot be attributed to one volunteer-hour.
Sources
Tended plants, garden or balconyNot quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Sources
Helped wildlife (feeder, bee hotel, a wild corner)Not quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Sources
Had a real conversation about climateNot quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Sources
Contacted a representative, signed, or showed upNot quantified
NatureNot quantified
Estimate
Not quantified
Likely range
No number shown
Compared with
not quantified
Sources

Sources

55 published sources behind the factors. Each name opens the publisher's page.

2026 Government greenhouse gas conversion factors for company reporting: Methodology paper (opens the publisher's page in a new tab)
UK Department for Energy Security and Net Zero · 2026
Used forWorked from home instead of commuting by car Took the train instead of a short flight Took the train instead of a short flight (distance unknown)
+1 more
Heating 1 °C lower
Also cited by 1 starting-line question.
Akenji et al. (2021) 1.5-Degree Lifestyles: Towards A Fair Consumption Space for All (Table 3.1, Figures 3.1-3.3) (opens the publisher's page in a new tab)
Hot or Cool Institute, Berlin · 2021
Also cited by 1 starting-line question, 1 equivalence, 1 lesson.
Apple (2025) iPhone 17 Product Environmental Report (opens the publisher's page in a new tab)
Apple · 2025
Used forRepaired instead of replacing Keeping my phone another year
Also cited by 1 lesson, 2 facts and posts.
Apple (2025) MacBook Air (M4) Product Environmental Report (opens the publisher's page in a new tab)
Apple · 2025
Used forRepaired instead of replacing
CO2 emissions per capita (Global Carbon Budget 2025, processed by Our World in Data) (opens the publisher's page in a new tab)
Our World in Data · 2025
Also cited by 1 equivalence.
Documentation for the Waste Reduction Model (WARM) v16: Containers, Packaging, and Non-Durable Good Materials (Exhibits 1-4, 2-7, 3-7, 5-3) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2023
Used forRecycled aluminium cans Recycled glass bottles or jars Recycled plastic bottles
+2 more
Recycled paper or cardboard Refilled my bottle
Documentation for the Waste Reduction Model (WARM) v16: Organic Materials (Exhibits 1-10, 1-43, 1-46) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2023
Used forSaved food from the bin Rescued a meal (leftovers, food-rescue app)Composted food scraps
Also cited by 2 lessons.
Edwards & Fry (2011) Life cycle assessment of supermarket carrier bags: a review of the bags available in 2006 (Report SC030148) (opens the publisher's page in a new tab)
Environment Agency (England) · 2011
Used forBrought my own bag
Also cited by 2 facts and posts.
Ember Yearly Electricity Data (full release, long format) (opens the publisher's page in a new tab)
Ember · 2026
Used forRode my e-bike or e-scooter instead of driving Drove electric instead of petrol Air-dried a load
+6 more
Switched standby devices off at the wall Replaced an old bulb with an LED Shorter shower Saved hot water Washed at 30 °C instead of 40 °C Washed cold instead of 40 °C
Also cited by 1 starting-line question, 2 equivalences, 2 lessons, 1 fact or post, 1 chart.
Energy Saving Trust - Lighting (opens the publisher's page in a new tab)
Energy Saving Trust · 2026
Used forReplaced an old bulb with an LED
Energy Saving Trust - Our calculations: fuel prices and carbon factors (updated 22 Sep 2026) (opens the publisher's page in a new tab)
Energy Saving Trust · 2026
Used forHeating 1 °C lower Switched standby devices off at the wall
Energy Saving Trust - Quick tips to save energy (updated 28 Sep 2026) (opens the publisher's page in a new tab)
Energy Saving Trust · 2026
Used forAir-dried a load Switched standby devices off at the wall Shorter shower
Energy Saving Trust - Saving water at home (opens the publisher's page in a new tab)
Energy Saving Trust · 2026
Used forHeating 1 °C lower Tap off while brushing
Energy Saving Trust - Thermostats and heating controls (opens the publisher's page in a new tab)
Energy Saving Trust · 2026
Used forHeating 1 °C lower
ENERGY STAR Clothes Dryers Key Product Criteria (opens the publisher's page in a new tab)
US EPA ENERGY STAR · 2014
Used forAir-dried a load
Foteinis, S. (2020) How small daily choices play a huge role in climate change: The disposable paper cup environmental bane. Journal of Cleaner Production 255:120294 (abstract) (opens the publisher's page in a new tab)
Journal of Cleaner Production · 2020
Used forUsed my own cup
Friedlingstein et al. (2026) Global Carbon Budget 2025. Earth System Science Data 18:3211-3288 (opens the publisher's page in a new tab)
Global Carbon Project / ESSD · 2026
GHG Emission Factors Hub 2025, Table 10 (business travel and employee commuting) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2025
Used forTook the bus instead of driving Took the train or metro instead of driving
Gifford et al. (2012) Residential Lighting End-Use Consumption Study: Estimation Framework and Initial Estimates (PNNL-22182) (opens the publisher's page in a new tab)
US Department of Energy / Pacific Northwest National Laboratory · 2012
Used forReplaced an old bulb with an LED
Greenhouse Gas Equivalencies Calculator - Calculations and References (page updated 4 Aug 2026) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2026
Used forWalked or cycled instead of driving Planted a tree
Also cited by 2 starting-line questions, 1 equivalence.
Hollingsworth, Copeland & Johnson (2019) Are e-scooters polluters? The environmental impacts of shared dockless electric scooters. Environ. Res. Lett. 14:084031 (opens the publisher's page in a new tab)
Environmental Research Letters · 2019
Used forRode my e-bike or e-scooter instead of driving
Household Electricity Survey: A study of domestic electrical product usage (Intertek report R66141 for DEFRA, DECC and the Energy Saving Trust) (opens the publisher's page in a new tab)
UK Government (DEFRA / DECC / Energy Saving Trust) · 2012
Used forAir-dried a load Switched standby devices off at the wall Washed at 30 °C instead of 40 °C
+1 more
Washed cold instead of 40 °C
Also cited by 1 equivalence.
IEA (2020) Working from home can save energy and reduce emissions. But how much? (commentary) (opens the publisher's page in a new tab)
International Energy Agency · 2020
Used forWorked from home instead of commuting by car
IEA & European Commission (2022) Playing my part: How to save money, reduce reliance on Russian energy, support Ukraine and help the planet (opens the publisher's page in a new tab)
International Energy Agency / European Commission · 2022
Used forHeating 1 °C lower
Also cited by 1 lesson, 1 fact or post.
India GHG Program (2015) India Specific Rail Transport Emission Factors for Passenger Travel and Material Transport, version 1.0 (opens the publisher's page in a new tab)
India GHG Program · 2015
Used forTook the train or metro instead of driving
India GHG Program (2015) India Specific Road Transport Emission Factors, version 1.0 (WRI India, TERI, CII) (opens the publisher's page in a new tab)
India GHG Program · 2015
Used forWalked or cycled instead of driving Took the bus instead of driving
Also cited by 1 starting-line question.
IPCC (2022) Climate Change 2022: Mitigation of Climate Change. Working Group III, Summary for Policymakers (statements B.3.4, C.10, C.10.2) (opens the publisher's page in a new tab)
Intergovernmental Panel on Climate Change · 2022
Also cited by 2 lessons, 4 facts and posts.
Levi Strauss & Co. (2015) The Life Cycle of a Jean: Understanding the environmental impact of a pair of Levi's 501 jeans (opens the publisher's page in a new tab)
Levi Strauss & Co. · 2015
Used forBought jeans second-hand Repaired instead of replacing
Also cited by 1 lesson.
McQueen, MacArthur & Cherry (2019) The E-Bike Potential: Estimating the Effect of E-Bikes on Person Miles Travelled and Greenhouse Gas Emissions (TREC white paper) (opens the publisher's page in a new tab)
Portland State University, Transportation Research and Education Center · 2019
Used forRode my e-bike or e-scooter instead of driving
Also cited by 1 starting-line question.
Method for Calculating Carbon Sequestration by Trees in Urban and Suburban Settings (Table 2) (opens the publisher's page in a new tab)
US Department of Energy, Energy Information Administration · 1998
Used forPlanted a tree
Also cited by 1 fact or post.
National Travel Survey England 2024: trips by purpose, age, mode and sex (published 27 Aug 2025) (opens the publisher's page in a new tab)
UK Department for Transport · 2025
Used forWorked from home instead of commuting by car
Also cited by 1 starting-line question.
NIH Environmental Management System - PET water bottle environmental impact (compilation of Benavides et al. 2018, PlasticsEurope 2017, ACC 2011, Franklin Associates 2009, Gironi & Piemonte 2010) (opens the publisher's page in a new tab)
US National Institutes of Health · 2019
Used forRefilled my bottle
Ofgem (2023) Decision on revised Typical Domestic Consumption Values for gas and electricity (opens the publisher's page in a new tab)
Ofgem · 2023
Used forHeating 1 °C lower
Also cited by 1 starting-line question.
ONS - Families and households in the UK: 2024 (average household size 2.35) (opens the publisher's page in a new tab)
UK Office for National Statistics · 2025
Also cited by 1 starting-line question.
Per capita consumption-based CO2 emissions (Global Carbon Budget 2025, processed by Our World in Data) (opens the publisher's page in a new tab)
Our World in Data · 2025
Per capita greenhouse gas emissions incl. land use (Jones et al. 2025, processed by Our World in Data) (opens the publisher's page in a new tab)
Our World in Data · 2025
Petrillo et al. (2018) Design of a sustainable electric pedal-assisted bike: a life cycle assessment application in Italy (IntechOpen) (opens the publisher's page in a new tab)
IntechOpen · 2018
Used forRode my e-bike or e-scooter instead of driving
Poore & Nemecek (2018) - Environmental footprints of dairy and plant-based milks, via Our World in Data (opens the publisher's page in a new tab)
Our World in Data / Science · 2018
Used forPlant milk instead of dairy
Poore & Nemecek (2018) - Food: greenhouse gas emissions across the supply chain, via Our World in Data (opens the publisher's page in a new tab)
Our World in Data / Science · 2018
Used forPlant-based instead of a beef meal Chose local, seasonal produce
Poore & Nemecek (2018) - GHG per 100 g protein, via Our World in Data (opens the publisher's page in a new tab)
Our World in Data / Science · 2018
Used forPlant-based instead of a beef meal
Poore & Nemecek (2018), Reducing food's environmental impacts through producers and consumers (Science 360:987-992) - GHG per kg, via Our World in Data (opens the publisher's page in a new tab)
Our World in Data / Science · 2018
Used forPlant-based meal Plant-based instead of a beef meal Chicken instead of beef
Also cited by 1 equivalence, 1 lesson, 3 facts and posts.
Ritchie, H. (2020) Very little of global food is transported by air; this greatly reduces the climate benefits of eating local (opens the publisher's page in a new tab)
Our World in Data · 2020
Used forChose local, seasonal produce
Ritchie, H. (2020) You want to reduce the carbon footprint of your food? Focus on what you eat, not whether your food is local (opens the publisher's page in a new tab)
Our World in Data · 2020
Used forChose local, seasonal produce
Also cited by 1 lesson, 2 facts and posts.
Scarborough et al. (2023) Vegans, vegetarians, fish-eaters and meat-eaters in the UK show discrepant environmental impacts. Nature Food 4:565-574 (Table 3) (opens the publisher's page in a new tab)
Nature Food · 2023
Used forPlant-based meal Vegetarian day Fully plant-based day
+1 more
Chose local, seasonal produce
Also cited by 1 starting-line question.
Specific heat capacity of liquid water, about 4.18 kJ/(kg.K) (NIST Chemistry WebBook, fluid properties of water) (opens the publisher's page in a new tab)
US National Institute of Standards and Technology · 2023
Used forShorter shower Saved hot water
Also cited by 2 equivalences.
UK Government GHG Conversion Factors for Company Reporting 2026 (flat file v1.2, DESNZ) (opens the publisher's page in a new tab)
UK Department for Energy Security and Net Zero · 2026
Used forWalked or cycled instead of driving Took the bus instead of driving Took the train or metro instead of driving
+21 more
Shared the ride Rode my e-bike or e-scooter instead of driving Drove electric instead of petrol Worked from home instead of commuting by car Took the train instead of a short flight Took the train instead of a short flight (distance unknown)Saved food from the bin Heating 1 °C lower Shorter shower Saved hot water Tap off while brushing Bought a top second-hand Bought jeans second-hand Repaired instead of replacing Borrowed or rented instead of buying Recycled aluminium cans Recycled glass bottles or jars Recycled plastic bottles Recycled paper or cardboard Composted food scraps Refilled my bottle
Also cited by 3 starting-line questions, 3 equivalences, 2 lessons, 2 facts and posts.
US Census Bureau QuickFacts - Persons per household, 2020-2024 (2.53) (opens the publisher's page in a new tab)
US Census Bureau · 2025
Also cited by 1 starting-line question.
US Department of Energy, Energy Saver - LED Lighting (opens the publisher's page in a new tab)
US Department of Energy · 2024
Used forReplaced an old bulb with an LED
WaterSense - Showerheads (updated 3 Mar 2026) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2026
Used forShorter shower
Also cited by 1 equivalence.
WaterSense - Statistics and Facts (updated 11 Mar 2026) (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2026
Used forTap off while brushing
Which? - Washing machine temperature guide (updated 9 Jul 2026) (opens the publisher's page in a new tab)
Which? · 2026
Used forWashed at 30 °C instead of 40 °C Washed cold instead of 40 °C
WRAP (2025) Displacement Rates Untangled - press release "Fast fashion could be left on the peg as preloved and repair displace new sales" (opens the publisher's page in a new tab)
WRAP · 2025
Used forBought a top second-hand Bought jeans second-hand Repaired instead of replacing
WRAP (2025) Household Food and Drink Waste in the United Kingdom 2022 (opens the publisher's page in a new tab)
WRAP · 2025
Used forSaved food from the bin Rescued a meal (leftovers, food-rescue app)
WRAP guidance: Expressing redistributed food surplus as meal equivalents (420 g per meal) (opens the publisher's page in a new tab)
WRAP · 2020
Used forRescued a meal (leftovers, food-rescue app)
Wynes & Nicholas (2017) The climate mitigation gap: education and government recommendations miss the most effective individual actions. Environ. Res. Lett. 12:074024 (opens the publisher's page in a new tab)
Environmental Research Letters · 2017
Used forWalked or cycled instead of driving Fully plant-based day Air-dried a load
+1 more
Washed at 30 °C instead of 40 °C
Also cited by 1 starting-line question, 2 lessons, 2 facts and posts.

Sources behind Learn, Impact and community postsLessons, facts and the charts on Impact cite these as well.38 SOURCES

2025 set to be second or third warmest year on record, continuing exceptionally high warming trend (opens the publisher's page in a new tab)
World Meteorological Organization · 2025
Aluminium recycling saves 95% of the energy needed for primary aluminium production (opens the publisher's page in a new tab)
International Aluminium Institute · 2024
Annual CO2 emissions (Global Carbon Budget 2025, processed by Our World in Data) (opens the publisher's page in a new tab)
Global Carbon Project / Our World in Data · 2025
Best before and use by dates (opens the publisher's page in a new tab)
UK Food Standards Agency · 2025
Climate Change 2007: The Physical Science Basis, FAQ 1.3: What is the greenhouse effect? (opens the publisher's page in a new tab)
IPCC Working Group I · 2007
Climate Change 2021, Chapter 7: The Earth’s energy budget, climate feedbacks and climate sensitivity (Table 7.15) (opens the publisher's page in a new tab)
IPCC Working Group I · 2021
Climate Change 2021: The Physical Science Basis, Summary for Policymakers (opens the publisher's page in a new tab)
IPCC Working Group I · 2021
Climate Equality: A planet for the 99% (opens the publisher's page in a new tab)
Oxfam and Stockholm Environment Institute · 2023
Clothes washers (opens the publisher's page in a new tab)
ENERGY STAR · 2025
Emissions by sector: where do greenhouse gases come from? (opens the publisher's page in a new tab)
Our World in Data · 2024
Emissions Gap Report 2025: Off target (opens the publisher's page in a new tab)
UN Environment Programme · 2025
Energy use in EU households in 2022 lowest since 2016 (opens the publisher's page in a new tab)
Eurostat · 2024
Food Waste Index Report 2024: Think Eat Save (opens the publisher's page in a new tab)
UN Environment Programme and WRAP · 2024
FOTW #1230: More than half of all daily trips were less than three miles in 2021 (opens the publisher's page in a new tab)
US Department of Energy · 2022
GISS Surface Temperature Analysis (GISTEMP v4), Land-Ocean Global Means (opens the publisher's page in a new tab)
NASA Goddard Institute for Space Studies · 2026
Global Carbon Budget 2024. Earth System Science Data 17:965-1039 (opens the publisher's page in a new tab)
Friedlingstein et al. · 2025
Global E-waste Monitor 2024 (opens the publisher's page in a new tab)
UNITAR and ITU · 2024
Global Plastics Outlook: plastic pollution is growing relentlessly as waste management and recycling fall short (opens the publisher's page in a new tab)
OECD · 2022
Globally representative evidence on the actual and perceived support for climate action. Nature Climate Change 14:253-259 (opens the publisher's page in a new tab)
Andre, Boneva, Chopra and Falk · 2024
Installed solar energy capacity (IRENA, via Our World in Data) (opens the publisher's page in a new tab)
IRENA / Our World in Data · 2026
Land use for agriculture: how much of the world’s land is used for food? (opens the publisher's page in a new tab)
Our World in Data · 2024
LED Lighting (opens the publisher's page in a new tab)
US Department of Energy · 2024
Life-cycle greenhouse gas emissions of combustion engine and electric cars in Europe, 2025 update (opens the publisher's page in a new tab)
International Council on Clean Transportation · 2025
Per capita CO2 and greenhouse gas emissions (Global Carbon Budget 2025; Jones et al. 2025) (opens the publisher's page in a new tab)
Global Carbon Project / Our World in Data · 2025
Quantifying aviation’s contribution to global warming. Environmental Research Letters 16:104027 (opens the publisher's page in a new tab)
Klöwer et al. · 2021
Renewable Power Generation Costs in 2024 (opens the publisher's page in a new tab)
IRENA · 2025
Scientific Assessment of Ozone Depletion: 2022, Executive Summary (opens the publisher's page in a new tab)
WMO / UNEP / NOAA / NASA / European Commission · 2022
Share of new cars that are electric (IEA Global EV Outlook 2026, via Our World in Data) (opens the publisher's page in a new tab)
IEA / Our World in Data · 2026
Solar photovoltaic module price (IRENA, via Our World in Data) (opens the publisher's page in a new tab)
IRENA / Our World in Data · 2025
Spending at least 120 minutes a week in nature is associated with good health and wellbeing. Scientific Reports 9:7730 (opens the publisher's page in a new tab)
White et al. · 2019
Standby power (opens the publisher's page in a new tab)
Lawrence Berkeley National Laboratory · 2025
The carbon footprint of streaming video: fact-checking the headlines (opens the publisher's page in a new tab)
IEA · 2020
The Future of Heat Pumps: how a heat pump works (opens the publisher's page in a new tab)
IEA · 2022
Trends in Atmospheric Carbon Dioxide: Mauna Loa annual mean (opens the publisher's page in a new tab)
NOAA Global Monitoring Laboratory · 2026
Used lithium-ion batteries (opens the publisher's page in a new tab)
US Environmental Protection Agency · 2025
What is a circular economy? (opens the publisher's page in a new tab)
Ellen MacArthur Foundation · 2024
What is ocean acidification? (opens the publisher's page in a new tab)
NOAA Pacific Marine Environmental Laboratory · 2024
What share of global CO2 emissions come from aviation? (opens the publisher's page in a new tab)
Our World in Data · 2024

The other numbers

The starting line, estimates for things we did not list, equivalences, and the world figures on Impact.

The starting line and your paceA one-minute lifestyle estimate, and a projection that is not your new footprint.

The optional starting-line quiz asks 6 questions about food, getting around, flights, home energy and shopping. It adds up published per-diet, per-mode and per-home values, and shows tonnes of CO₂e a year to one decimal place (model version 1). Treat it as accurate to about ±40% for a typical person, and worse at the extremes.

It covers the parts of a footprint you steer day to day. It leaves out public services, infrastructure, leisure and services, which together are about a quarter to a third of a full footprint. That is why it is not comparable with national per-person figures.

Your pace compares what you have logged over the last four weeks, projected to a year, with your starting line. It needs 14 days since your first log and at least 7 active days before it shows anything. It counts only recurring swaps beyond the habits you told us you already had, and one-off actions are added once, never multiplied.

Custom actions and AI estimatesEstimates for things we did not list are labelled and kept out of your headline total.

For something that is not in the catalogue, Touch Grass first looks for a close match. If there is none and the live AI is connected, the model gives a conservative guess for the action you typed. Otherwise you choose a category and an effort level and no CO₂e is claimed.

AI estimates are always marked as low confidence and carry a dotted underline wherever they appear.
They are capped at 2 kg per log and 5 kg per day.
They are kept out of the headline total, the share card and the pace metric. The total shows them on a separate line.

EquivalencesComparisons are meant to give scale, not to describe outcomes.

The Impact page translates your total into everyday units, such as kilometres in an average car. There are 8 in the catalogue, and each is phrased "That is roughly the CO₂ from…". It is a way to feel the size of a number, never a claim about what you achieved.

We do not turn your total into "trees". In an app with a tree as its mascot, "that is 3 trees" would sound like trees planted, which is not true.

The world figures on ImpactLive readings from public sources, shown beside your own numbers and never added to them.

The world view of the Impact page shows public climate data. Some charts are datasets bundled with the app, so they work offline; they are listed under Sources. Five readings are fetched live: global temperature, CO2, methane and nitrous oxide in the air, and CO2 per person. They come from the publishers named below, through our own server, so your browser never contacts them.

The server keeps each answer for a day and checks it before use: a reading that is the wrong shape, outside a physically sensible range or too old is treated as if the source were down. When that happens, or when you are offline, the page shows a snapshot saved with the app, labelled with its date. The "Live" badge appears only on a reading fetched within the last day and a half.

These figures give context. They are not used in any of your totals, ranges or equivalences.

Global surface temperature
Monthly change against the 1951 to 1980 average, land and ocean.
GISTEMP v4, NASA Goddard Institute for Space Studies
Carbon dioxide in the air
Global daily average in parts per million, with and without the seasons.
NOAA Global Monitoring Laboratory
Methane and nitrous oxide in the air
Global monthly averages in parts per billion.
NOAA Global Monitoring Laboratory
CO2 per person
Territorial emissions per person for the world and your region, without land-use change. The latest year each country has reported.
World Bank, World Development Indicators (EN.GHG.CO2.PC.CE.AR5)

What we do not claim

The honest limits of a self-reported estimate, and the habits that keep it honest.

We can claim

Yes:An approximate amount of CO2e avoided by one action, compared with a stated alternative.
Yes:Which actions are large (flights, car travel, beef, heating) and which are small (bags, cups, standby, tap water).
Yes:A running personal total, labelled "estimated CO2e avoided".

We cannot claim

No:That your footprint fell by exactly this amount. Nobody measured what you would otherwise have done.
No:That a personal total "offsets", "neutralises" or "cancels" any emissions.
No:That figures apply everywhere. Most transport, waste and home-energy factors are UK or US values.
No:Any CO2e benefit for litter picking, volunteering or other actions that are not quantified.
No:That planting a tree removes carbon today, or permanently.

Counting each thing onceOverlapping actions and daily caps stop one habit being counted twice.

Some actions describe the same thing in different words. The Log screen applies caps and overlap groups to keep the total honest. These are the rules behind them, and a good habit to follow even without the caps.

One trip, one action: do not log "walked instead of driving" and "took the bus instead of driving" for the same journey.
Diet: log a vegetarian or vegan day, or individual plant-based meals, not both. "Chicken instead of beef" and "plant-based meal" overlap too.
Food waste: "food waste avoided (kg)" and "meal saved" are the same thing in two units. Composting applies only to scraps that were not eaten.
Laundry: one load is either 30 °C or cold, plus optionally line-dried.
Hot water: "shorter shower" and "hot water saved" overlap.
Carpooling: the saving belongs to the car. Share it between the riders, do not give it to each.
Household actions such as the thermostat and standby: one log per household per day.
Recycling versus buying recycled: the benefit is credited to the recycler here, so do not also credit recycled-content purchases.
Recycling a bottle and refusing a bottle are alternatives for the same bottle.
Second-hand and repair use displacement rates, because not every purchase or repair replaces a new item.
One-off actions such as an LED bulb, a tree or keeping a phone count per day or per year going forward. They are never a lump sum and never back-dated.
The starting-line quiz and the action log use different methods. Subtracting one from the other gives an indication, not a new footprint.

Known gapsWhy ranges are wide, and what this release does not model.

The alternative is unknown: which car, how full the bus, what you would have eaten.
Electricity varies about 15-fold between countries (France 41 against India 671 gCO₂e per kWh in 2025) and by the hour.
Farm-level food footprints vary several-fold for the same product. Beef from a dairy herd and beef from a beef herd differ by a factor of three.
Method choices matter: with or without the non-CO₂ effects of flying, with or without land-use change, average or marginal electricity, who gets credit for recycling.
Some studies are old or narrow. The carrier-bag study uses 2006 UK data, the appliance measurements come from 251 English homes in 2010 and 2011, and the India transport factors date from 2015.
Item sizes are assumed: a can, a bottle, a T-shirt, a meal.

Not modelled in this release: hourly or live grid intensity, your own vehicle or energy bills, and any CO₂e claim for air-conditioning. Where we use annual averages or three regional car factors, the number is an approximation of your real situation.

How we word numbers

The writing rules the whole app follows.

Always

Yes:Put an approximately-equal sign or the word "about" in front of every number.
Yes:Say "estimated CO2e avoided", never "saved the planet", "offset" or "removed".
Yes:Name the comparison: "vs driving an average car".
Yes:Link every figure to this page and its sources.
Yes:Round honestly: two significant figures at most ("about 1.5 kg", "about 20 g").
Yes:Use grams below 0.1 kg, so tiny actions look tiny.

Never

No:"You saved X kg of CO2" as a bare fact.
No:"Carbon neutral", "net zero", "offset" or "cancelled out".
No:More than two significant figures.
No:A CO2e number for an action that is not quantified.
No:Comparisons that shame, such as "you emitted…".
No:Equivalences presented as outcomes, such as "you planted 3 trees".

Versions and updates

Which factors are in use, how they are rounded and when they change.

Factor version 2026.10, built on 2026-10-06. Central values are rounded to two significant figures, ranges likewise. Full precision is kept internally and a figure is shown to two significant figures at most.

CO₂e uses 100-year global warming potentials as embedded in each source: the UK 2026 factors and the US EPA use IPCC AR5, Poore and Nemecek use AR5 with climate-carbon feedbacks, and Apple uses AR6. We do not re-harmonise them. The differences are small next to the other uncertainties.

Factors are refreshed after each annual release: the UK conversion factors in June, Ember in spring, the Global Carbon Budget in November, and NASA and NOAA in January. New logs use the new version. Old logs keep the version they were saved with.

Changelog

2026.102026-10-06
- First release: 51 actions in 7 categories, 45 grid regions, 8 equivalences and the starting-line model version 1.
- Factors built from the UK Government 2026 conversion factors, Ember 2025 grid data, Poore and Nemecek 2018, US EPA and WRAP sources, and manufacturer life-cycle reports.

Next: what Touch Grass stores and what leaves your device.