How efficient is an approach of geoengineering to mitigate the - - PowerPoint PPT Presentation

How efficient is an approach of geoengineering to mitigate the global warming?

I.I. Mokhov, A.V. Eliseev, and A.V. Chernokulsky A.M. Obukhov Institute of Atmospheric Physics RAS

ENVIROMIS-2008

adopted from [IPCC, 2007]

Motivation - 1

• Globally, climate has warmed in the 20th century by 0.6 K

(0.4-0.8 K).

• Such warming on a century timescale was not observed for

any previous epoch and most likely to be attributed to human activities [IPCC, 2007]. This warming is expected to proceed for the whole 21st century and beyond.

Motivation - 2

It was suggested [Budyko, 1974] to mitigate global warming by injection of sulphur in the stratosphere. Recently, this approach is considered as a form of geoengineering [Schneider 1996; Schneider 2001; Izrael, 2005; Crutzen, 2006; Wigley, 2006]. Natural examples: cooling after volcanic eruptions; less direct: cooling due tropospheric sulphates Benefits: low cost due to large residence time of aerosols in stratosphere (~2-3 yr) Possible disadvantages: strong decrease of precipitation [Trenberth and Dai, 2007] possible enhancement of the stratospheric ozone depletion [Tilmes et al., 2008] Emissions required to compensate the atmopsheric CO2 doubling: [Izrael, 2005]: 0.6 TgS/ yr [Crutzen, 2006]: 1-2 TgS/ yr [Wigley, 2006]: 5 TgS/ yr

IAP RAS CM

Resolution: 4.5o*6o, L8 - atmosphere, L4 - ocean, L1 - land; ∆t = 5 days Atmosphere: 3D quasigeostrophic large- scale dynamics. Synoptic- scale dynamics is parametrised based on their representation as Gaussian ensembles. In any atmospheric layer, temperature depends linearly on height. Fully interactive hydrological cycle. Partly interactive methane cycle. Ocean: Prognostic equation for sea surface temperature. Geostrophic large- scale dynamics. Universal vertical profiles in any oceanic layer. Oceanic salinity is prescribed. Interactive, globally averaged oceanic carbon cycle. Sea ice: Diagnostic, based on the local SST Vegetation: Spatial distribution of ecozones is prescribed. Fully interactive globally averaged terrestrial carbon cycle. Interactive CH4 emissions from natural wetlands. Turnaround time: ~ 17 sec per model year (Intel Zeon)

Top-of-the atmosphere stratospheric aerosol radiative forcing

Fstrat = - astrat τstrat, astrat = 22 W/ m2 [Hansen et al, 2005],

• ptical depth

τstrat = kext,strat Mstrat Mstrat is stratospheric aerosol mass per unit area, extinction coefficient kext,strat = 7.6 m2/ g (derived from the Mt. Pinatubo A.D. 1991 eruption observations)

Annual mean surface air temperature [K] response to volcanic forcing [Amman et al., 2003] (1891-2000)

A.D. 1992 (aftermath for Mt.Pinatubo eruption)

IAP RAS CM

• bs., ENSO removed

[Wigley, 2000]

A g u n g

E l C h i c h

• n

M t . P i n a t u b

• ∆Tg, K

Ensemble numerical experiments with a climate mitigation via stratospheric aerosol loading

• duration: 1860-2100
• historical+SRES A1B anthropogenic CO2 and CH4

emissions

• historical+SRES A1B atmospheric concentrations of

N2O (BernCC) and tropospheric sulphates (MOZART 2.0) + mitigation via controlled sulphur emissions in the stratosphere with values of governing parameters varying between different ensemble members The total number of ensemble members: 2331 Cumulative length: 564 102 yr

Parameters of st ratospheric aerosols

Global burden: d Mstrat,g / d t = E - Mstrat,g / τres Emissions: 0, before A.D. 2012 E = E

0, from A.D. 2015 to t 0

0, after t 0 Local burden: Mstrat = ( Mstrat,g / S

Earth ) * Y(φ)

Depending on the ensemble member E

0 = from 0.6 to 4 TgS/ yr

t 0 = A.D. 2100 or A.D. 2075 kext, strat = 5-20 m2/ g residence time τres = 1-4 yr latitudinal profile Y(φ) is varied between uniform, triangular, and trapezoidal functions of x = sin φ with varying either x 0 or x 1 (see Figure).

{

NP x 0 x1 EQ

• x 1

SP

x Y

Change in global surface air temperature

whole ensemble ensemble members with τres=2 yr, kext,strat=7.6 m2/g, and uniform Y(φ)

no mitigation E

0 = 0.6 TgS/ yr

E

0 = 1 TgS/ yr

E

0 = 2 TgS/ yr

E

0 = 3 TgS/ yr

E

0 = 4 TgS/ yr

• bs. (CRU UEA)

∆Tg, K ∆Tg, K

Mitigat ion eff iciency of diff erent latitudinal

profiles for stratospheric aerosol Y(sin φ)

∆Tmitigation,g-∆Tanthrop,g in year 2100, K τstrat,* uniform triangular with φ0=70oN trapezoidal with φ1=50oN triangular with φ0=30oN trapezoidal with φ1=30oN

τstrat,* = E

0 τres kext,strat/ S Earth

E

0 - emissions of

stratospheric aerosols τres - residence time kext,strat - extinction coefficient S

Earth - area of the Earth's

surface

Spat ial pattern of mit igation eff iciency for 2050-2060:

• (∆Tm itigation-∆Tanthrop)/ (∆Tmitigation,g-∆Tanthrop,g)

Tm itigation - ensemble members with climate mitigation Tanthrop - ensemble member without mitigation

uniform Y(sin φ) triangular Y(sin φ) with φ0=70oN

Global precipitation change

whole ensemble ensemble members with τres=2 yr, kext,strat=7.6 m2/g, and uniform Y(φ)

no mitigation E

0 = 0.6 TgS/ yr

E

0 = 1 TgS/ yr

E

0 = 2 TgS/ yr

E

0 = 3 TgS/ yr

E

0 = 4 TgS/ yr

∆P

g, %

∆P

g, %

Pattern of relative precipitat ion response to mitigation for 2050-2060: 100*(∆Pm itigation-∆Panthrop)/ P0

Pmitigation - ensemble members with climate mitigation Panthrop - ensemble member without mitigation P0 - present-day annual precipitation

uniform Y(sin φ) triangular Y(sin φ) with φ0=70oN

Change in global surface air temperature in experiments with a mitigation emission stop in 2075

∆Tg, K dTg/dt, K/yr no mitigation E

0 = 2 TgS/ yr

• bs. (CRU UEA)

SAT change rate [K/ decade] for 2076-2085 after the mit igation stop in 2075 (E0 = 2 TgS/ yr)

without mitigation triangular Y(sin φ) with φ0=70oN, τres=2.5, kext,strat=7.6 m2/ g triangular Y(sin φ) with φ0=70oN, τres=4, kext,strat=20 m2/ g

Globally averaged energy-balance model

C dTg / dt = Q [ 1 - α(Tg) ] - ( A + B Tg )η + Fstrat,g, C - heat capacity per unit area, Tg - globally averaged surface air temperature, t - time, Q - insolation, α - planetary albedo, A and B - constants, correction factor for anthropogenic greenhouse effect η = 1- c0 log (qC/ qC,0), c0 = 2.3*10-2, qC (qC,0) is the current (initial) atmospheric CO2 concentration. Equilibrium climate sensitivity to CO2 doubling in the atmosphere [Mokhov, 1981]: ∆T2CO2 = ( c0 I0 log 2 ) / ( 1 - Q kα + B) where I0 = A + B Tg,0 a nd ka = d α / dTg, subscript '0' indicates the present-day state

EBM forcings specification

i) qC = qC,0 exp ( t / t p) t p - prescribed time scale ii) Fstrat,g = - astrat τstrat,g, astrat = 22 W/ m2,, τstrat,g = kext,strat Mstrat,g and Mstrat,g = τlife E

0 = const

(stationarity approximation for Mstrat,g). Governing parameters are varied between the different ensemble members: E

0 = 0.6-5 TgS/ yr

τres = 1-4 yr kext,strat = 5-20 m2/ g ∆T2CO2 = 1.5-4.5 K t p = 50-250 yr-1

Global temperature change in years 0-100

• btained with an energy-balance model

no mitigation mitigation with τres=2 yr, kext,strat=7.6 m2/ g

E

0=1 TgS/ yr

E

0=4 TgS/ yr

tp, centuries ∆T2CO2, K tp, centuries tp, centuries ∆T , K IAP RAS CM

10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12

t p=100 yr (~SRES A2), τres=2 yr t p=100 yr (~SRES A2), τres=3 t p=136 yr (~SRES A1B), τres=2 t p=136 yr (~SRES A1B), τres=3 t p=230 yr (~SRES B1), τres=2 t p=230 yr (~SRES B1), τres=3

year E, TgS/yr

Emissions required to compensate the greenhouse-gases-induced warming (energy-balance model)

Conclusions - 1

• For large annual emissions, large residence time of sulphates in

the stratosphere, and large extinction coefficient it is possible to mitigate both global and regional warming to a large extent. However, if the ranges for above parameters are narrowed to presumably more realistic widths, the residual warming is > 1.8 K in the 21st century. Globally, the most efficient latitudinal distribution of geoengineering aerosols is that with high loading in the extratropics. At regional scale, other latitudinal distributions may be preferable.

• However, stratospheric aerosol climate mitigation leads to less

humidification of arid regions in comparison to non-mitigated anthropogenically induced warming. A caveat in this result is due to prescribed atmospheric relative humidity in the IAP RAS CM.

Conclusions - 2

• Due to the fast removal of the mitigation effect if the

corresponding emissions are stopped climate trajectory returns to the non-mitigated one within a few decades. This results in a necessity to continue mitigation very long in future, perhaps for several centuries in order to make it efficient.

• The results obtained with the IAP RAS CM are further supported

and interpreted by making use of an energy-balance climate

• model. It is shown that very high stratospheric sulphate emissions

(up to 12 TgS/yr) are needed to compensate global warming expected in the 21st century.

Thank you f or attention! Thank you f or attention!

General structure of the IAP RAS CM

ATMOSPHERE

insolation

OCEAN VEGETATION

(prescribed ecozones, interactive carbon cycle)

SOIL ICE SHEETS

(prescribed)

PERMAFROST

heat, moisture, momentum, CO2

SEA ICE

clouds (single effective layer)

strato- and mesosphere free troposphere boundary layer

large-scale circulation, synoptic-scale processes are parameterised heat, moisture, CO2

SNOW

run

• ff

litterfall

mixed layer

(thermal and hydrological processes, carbon cycle)

deep ocean bottom friction layer

in the whole ocean the following processes are considered: heat transport, large-scale circulation, synoptic-scale processes are parameterised, prescribed salinity precipitation shortwave radiation longwave radiation convection concentrations N2O, freons, tropospheric and stratospheric aerosols anthropogenic emissions СО2, CH4

Surface air temperature change (K) with respect to 1860

2000

• bs. UEA CRU,

(2005-1996)-(1896-1905) А1В, 2100 А2, 2050

Annual precipitation change (%) with respect to 1860

2000 А1В, 2050 г. А1В, 2100 В1, 2050

• bs. [Dai et al., 1997],

(1990-1995)-(1896-1905)