For solar constant and planetary albedo change , the global-mean absorbed shortwave perturbation is . The factor four comes from intercepted disk area versus total spherical area. The incoming top-of-atmosphere sunlight itself is unchanged.
The heat capacity per unit horizontal area is : mass density times specific heat capacity times atmospheric depth. Thus is the rate of stored thermal energy per unit area. The solar constant is the incident radiative flux on a surface normal to sunlight; the ratio of the Earth's intercepting disk area to its spherical area is . The absorbed fraction is , where albedo is the reflected fraction. The outgoing radiative flux is , a Stefan–Boltzmann law with effective radiation factor incorporating the simplified greenhouse effect. Every term therefore has units of power per area.
The albedo curve decreases smoothly, passes through , and approaches at high temperature. At it is , close to when is large. Its derivative is
An ice-covered planet reflects more sunlight than exposed land or ocean. Warming removes reflective ice, lowers albedo, and increases absorbed radiative flux: this is ice-albedo feedback. The parameter controls the sharpness of this transition; the formula is an idealized global response, rather than a local melting law.
Figure 1.
Temperature-dependent albedo and the separated fast cooling and slow recovery
. Details.
Balances and meanings. The first equation is conservation of energy per unit area. Here is effective areal heat capacity, is the solar constant, and accounts for the ratio of the Earth's intercepting disc area to its total surface area. The albedo reflects a fraction of that input. The outgoing thermal radiation is the Stefan–Boltzmann law modified by , an effective transmission factor for the greenhouse effect; it normally decreases as atmospheric carbon dioxide increases.
The albedo relaxes towards its temperature-dependent equilibrium on the ice-adjustment time . The atmospheric mass inventory per unit pressure is , using the molar mass ratio of carbon dioxide to air and hydrostatic atmospheric mass. Its sources and sinks are volcanic input , removal by chemical weathering over land area at rate , and the atmosphere-to-ocean exchange . The ocean equation is a molar mass conservation balance: is ocean water mass, division by converts the incoming carbon-dioxide mass to moles, and removes dissolved carbon by burial/export. With in moles per kilogram, has units of kilograms of water-equivalent per year.
Times and reduction. Linearize the temperature equation at fixed about radiative equilibrium. Its exponential relaxation time is
The equality uses . A stock-over-flux thermal time, sometimes used as a scale, is four times this, ; both describe a rapid thermal response relative to the other processes.
For pressure in atmospheres, write for the atmospheric carbon-dioxide mass per atmosphere. The factor is essential because is given per atmosphere. The pressure and slow ocean exponential relaxation times are
Thus . The leading quasi-steady approximation gives
Substitution in the two remaining equations yields
The coefficient of remains ; the printed in the reduced equation is a typographical error. Eliminating a fast temperature variable does not replace the ice-response time. The ocean time here is the slow time after atmospheric balance cancels the exchange flux; holding pressure fixed after imposing a carbon-dependent would describe a different conditional relaxation.
The supplied volcanic input gives
about of the stated present value ; that value is about .
Carbon-dioxide levels. Since , the atmospheric balance with gives
This uses the specified present ocean concentration, as requested, rather than replacing it by the model's smaller volcanic-only equilibrium.
After sustained additional input , barely changes during the atmospheric adjustment. At this stage
This is an intermediate balance, approached over several centuries. On the slower ocean time,
with pressure following . Since , pressure continues to rise slowly. Its eventual level in this stipulated constant-parameter model is
The two-reservoir carbon adjustment therefore has a rapid rise towards the frozen-ocean balance followed by a much longer increase. The figure solves the coupled linear pressure–ocean equations, retaining their small inventory correction to the reduced time scales.
Figure 1.
Atmospheric carbon dioxide adjustment over centuries and over the slow ocean-carbon time
. Details.
The solar constant measures the incident power per area normal to the sunlight. A spherical Earth of radius intercepts sunlight over the disc , but radiates over its surface . For uniform surface albedo in the global energy-budget approximation, absorbed power is . Treat the surface as a blackbody, assume efficient heat redistribution and neglect internal heating. The Stefan–Boltzmann law and the planetary surface energy balance then give
For a nonuniform surface the same power balance instead defines an effective planetary equilibrium temperature, ; its fourth root is not generally the arithmetic mean of the local temperatures. Also, area-averaging the surface albedo is an approximation to the properly insolation-weighted averaging. These are the simplifications implicit in the naked-planet calculation.
Take , so . The changed area is , giving
The initial planetary equilibrium temperature is . Thus
The differential estimate gives the same result. This is the global equilibrium response of the specified model, not the immediate local Arctic warming or the full atmospheric response.
The exceptional 2007 Arctic sea ice decline is best understood as a weather event acting on an already vulnerable ice cover, with several reinforcing pathways. Warmer air increases downward long-wave heating, lengthens the melt season and delays autumn freezing. Changes in snowfall and the development of melt ponds also change the surface energy balance. Air temperature alone, however, does not identify the source of the observed bottom melt: heat must reach the underside through the ocean.
Warmer Atlantic-origin water supplies a larger subsurface ocean heat content after entering through Fram Strait and the Barents Sea. Its influence on sea ice depends on vertical exchange, because a cold, relatively fresh halocline normally separates it from the ice. Erosion of that halocline, through altered freshwater distributions, wind-driven mixing or changes in circulation, weakens stable density stratification and can increase upward ocean heat flux. A warmer Atlantic layer without this exchange need not produce immediate extra bottom melt. Conversely, warmer Pacific inflow through Bering Strait introduces heat into the shallower western Arctic; advection through the Chukchi Sea provides another route to ice retreat and ocean heat storage.
The Arctic Oscillation changes winds, ice motion and export. Positive phases during earlier years helped remove old, thick ice through Fram Strait and left a younger, thinner pack. Summer 2007 also had a pronounced pressure contrast between the Beaufort region and the Eurasian Arctic, associated with the Arctic dipole anomaly. Its winds promoted ice displacement and warm-air transport. The summer circulation should not be described merely as a particular Arctic Oscillation index: the spatial pattern and the preconditioning from earlier years both matter.
Long-term thinning reduces the latent heat of fusion needed to remove the remaining ice. Replacement of multi-year sea ice by first-year sea ice therefore makes a given summer heat input more effective, while mobile thin ice is more readily redistributed or exported. These dynamical changes can expose open water early enough for substantial solar absorption. Water and melt ponds have lower surface albedo than snow-covered ice, so absorbed short-wave energy rises, increasing upper-ocean heat storage and bottom melt. This ice-ocean albedo feedback also delays refreezing. In the Beaufort Sea in 2007, enhanced solar heating of the newly exposed upper ocean was an especially important explanation of extreme bottom melt. Thus thinning and wind-driven opening can initiate retreat, while solar heating amplifies it; changes in Atlantic or Pacific heat supply can further reinforce the process where circulation permits.
A further positive climate feedback involves terrestrial carbon. Loss of the insulating ice cover permits greater autumn and winter transfer of stored ocean heat to the atmosphere. The resulting regional warming can increase thaw of coastal and inland permafrost. Microbial decomposition of previously frozen organic material then releases carbon dioxide, and oxygen-poor wet environments also release methane. These greenhouse gases strengthen the greenhouse effect, increasing global warming and further thaw. The permafrost carbon feedback provides the causal chain from summer ice retreat to an additional global influence. An Arctic methane feedback involving subsea sources is also possible, but the amount actually reaching the atmosphere depends on oxidation and transport; an abrupt, wholesale release is not a necessary consequence of summer ice loss.
For the bottom-melt estimate, use and the given . Assume the ice at the melting interface is already at its local freezing temperature; there is then no separate sensible-heating term. Per square metre the latent heat requirement is
July through September has days, so the mean ocean heat flux to the ice is
Using instead gives about , within the intended accuracy. Around the date of this examination, the estimated net anthropogenic radiative forcing since preindustrial times was about ; doubling carbon dioxide gives about . The calculated bottom flux is tens of times larger. There is no energy-budget contradiction: it is a local, summer ocean-to-ice heat transfer, supplied by absorbed sunshine, stored heat and transport, whereas radiative forcing is a global perturbation to the radiation budget. Redistribution and local absorption can produce much larger regional fluxes than the global forcing.
For an already ice-free water column, take and liquid-water specific heat capacity . Assuming all the same energy remains in that uniformly mixed column,
If the water initially lies near the seawater freezing point, about , its final temperature is approximately . An absolute final temperature requires that additional initial condition; the energy calculation itself determines only the temperature increase. The given ice specific heat capacity is not the appropriate value for heating liquid seawater.
Zero-dimensional planetary energy balance Created 2026-10-13 Updated 2026-10-20
Consider an idealized planet that is a sphere of radius , illuminated by parallel sunlight of irradiance , measured on a plane perpendicular to the rays. Its intercepting projected area is , whereas its total surface area is . Therefore the incoming power averaged per unit total surface area is
where is the broadband Bond albedo. The factor four is a sphere geometry result, not an atmospheric transmission factor. Local sunlight still varies over the illuminated hemisphere and vanishes on the dark side.
Now stipulate a single spatially uniform temperature , no atmosphere, no internal heating, and emissivity one for outgoing thermal radiation. This is a thermal blackbody model using the Stefan–Boltzmann law; the prescribed solar-band reflection is treated separately, so the body is not a perfect absorber at all wavelengths. Let be the effective heat capacity per unit surface area, in . Conservation of energy gives the ordinary differential equation
Here is net heating per unit surface area, in , and is the Stefan-Boltzmann constant. Use , rounded from the exact constant reported by NIST. [ref] The uniform temperature is a model assumption: generally the surface average of differs from the fourth power of the surface-average temperature.
For fixed Bond albedo , radiative equilibrium requires , giving
With the illustrative solar constant , the global incoming average is :
Fixed Bond albedoReflected average ()Absorbed average ()Equilibrium temperature ()
Doubling albedo here lowers the equilibrium temperature by , rather than halving it: the absorbed fraction changes from to , and thermal radiation scales as . These stipulated Bond albedos are whole-planet examples, not measurements of bare water and ice. The model predicts neither Earth's modern surface temperature nor its response to actual climate change; it omits the greenhouse effect, spatial heat transfer and seasons.
Figure 1.
From a sunlit disk to a whole sphere: geometry and fixed-albedo energy budgets
. The disk intercepts sunlight over area pi R squared; a uniform emitting sphere has four times that area. Both fixed-albedo examples use the same 1360 watts per square metre illumination and thermal emissivity one. Details.
Each fixed-albedo equilibrium is a stable equilibrium: for , the derivative is . A small temperature increase produces net cooling, and a small decrease produces net heating. This is the simplest expression of the stabilizing Planck feedback. Allowing albedo to fall with temperature adds a competing term, as in ice-ocean albedo feedback.