UPDATED briefest possible explanation of CO₂ #climatechange physics:
(1/12) Any warm object radiates energy at wavelengths depending on its temperature, the distribution of which is described by Planck's Law:
2. The sun's surface temperature is 5,778 K, so Planck's Law predicts it will radiate primarily at visible light wavelengths.
3. Earth receives 1,362 W/m² from the sun in a combination of UV, visible and infrared radiation, but 29.5% is immediately reflected by clouds or the surface. The remaining 70.5% (960 W/m²) is absorbed by land, air and oceans.
4. The sun illuminates the disk of Earth (area = πr²) but our spherical plant has an area = 4πr². To maintain energy equilibrium and steady surface temperature, the average square meter of Earth's surface must emit 240 W/m² (960 W/m² ÷ 4 = 240 W/m²).
5. From the Stefan-Boltzmann Law, we can calculate the surface temperature that will radiate exactly 240 W/m² (255 K). If we plug this temp. back into Planck's Law, we calculate that Earth will glow in a distribution of wavelengths mostly in thermal infrared (4-50 μm).
6. This would be the case if Earth had no atmosphere. Instead, greenhouse gasses (GHGs) absorb & re-radiate Earth's outgoing infrared, producing a jagged emission to space.
7. IR-active GHGs suppress emission in some portions of the spectrum, requiring surface warming of about +33 K to maintain 240 W/m² of outgoing thermal infrared (equilibrium with the sun). But the warmer surface now emits 390 W/m², 150 W/m² more than it would without GHGs.
8. This +33 K warmer surface temperature and the additional 150 W/m² thermal infrared is the greenhouse effect (GHE). Human additions of CO₂, CH₄, and other minor gasses have further increased this by ~3 W/m² so far.
9. How much warming does +3 W/m² produce?
•Easier part is calculating effects from infrared absorption alone.
•Harder part is how clouds, ice, and vegetation respond. Clouds and ice reflect a lot of sunlight, while trees absorb it. Climate models don't agree on these feedbacks.
10. Consequently, warming estimates have historically varied over a wide range: 1.0-6.0°C per doubling of CO₂ (Knutti-2017). Recently the IPCC AR6 report has concluded there is enough evidence to tighten this range to 2.5-4.0°C.
11. Future warming also depends on how quickly & how much more GHG's are added. While some have feared worst-case-scenarios RCP8.5 or SSP5-8.5, current emissions projections much more modest (dire media headlines are usually based on RCP8.5 studies)
12. Any campaign to change the status quo may tend to construct two simple, clear-cut categories: pro- and anti-climate action. In fact, Americans can be categorized according to at least six levels of concern regarding climate change:
As usual, my goal is 100% accuracy. If something is incorrect, feel free to let me know.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Let’s debunk another electric car meme that @TakesByLevin and @BJul1989 seem to think is true:
1) That isn’t a lithium mine, it is a diamond mine. The Mirny mine in Siberia, Russia closed in 2004.
https://t.co/crFyplhVdYamusingplanet.com/2013/04/abando…
2) US aviation fuel consumption is currently 23 billion gal/yr, close enough to claimed 19 billion gal/yr.
By now you've surely seen this figure, it was the talk of #climatetwitter today. But the most interesting part is the region I've circled. It's where Earth is constantly sliding in and out of ice ages 📉❄️📈🌲📉❄️📈🌲📉❄️
There's a paper you don't hear about very often Ganopolski-2016 that starts with Earth's orbital eccentricity, obliquity, and precession, which affect how much solar radiation the Northern hemisphere receives. Then it calculates...
...how much CO₂ is needed to boost the greenhouse effect enough to have prevented those ice ages. Turns out, ~425 ppm is pretty good estimate for "ice-age proofing" Earth.
Briefest possible explanation of CO₂ #climatechange physics - the DIY version:
(1/10) Earth receives 1,362 W/m² from the sun in a combination of UV, visible and infrared radiation.
📏You can measure it for yourself with a Hukseflux DR30-D1 Pyrheliometer
(2/10)The sun illuminates the disk of Earth (area = πr²) 📏You can calculate Earth's area for yourself the way the Greeks did it using shadows, or with a straight-edge and a view of an ocean + trig:
(3/10) 29.5% of incoming solar radiation is reflected by clouds or the surface.
📏You can measure this yourself with a Hukseflux SRA15 albedometer:
This graphic suggests that manufacturing one electric vehicle causes enormous diesel emissions from the earth mover alone.
Is it true? Let's check the math:
From Caterpillar's website the 994H has a 1577 HP engine and can lift 35 tons (77,000 lbs) at once.
1577 HP consumes 85 gal/hr at 100% throttle, but it's not expected to consistently operate at full throttle & rated load (more likely 40-70% of each, I will use 60%)
One Tesla 85 kWh battery has about 275 kg of aluminum, which requires scooping up 5 tons of the red mineral bauxite. The CAT 994H doesn't haul the material, only lifts it into the truck. Assuming 30 lifts/hr, it consumes ~2 gal diesel/lift.
1 lift = 21 tons (4.2 Tesla batteries)
I'm ~not~ saying the climate consensus is wrong. I'm explaining why we even ~talk~ about a consensus.
With no direct CO₂ forcing measurement, climatologists study indirect effects like 🌡️ ocean heat content or 📉 stratosphere instead.
The effect of aerosols, ozone, ⛅️ & 🌋 are all accounted for, then what's left over is then attributed to increased CO₂, even though CO₂ radiative forcing wasn't directly measured (it was modeled w/computers).