1 Climate forcings for C20C: What do we include and what should we include? Jeff Knight, Adam Scaife and Chris Folland Hadley Centre for Climate Prediction.

1

Climate forcings for C20C: What do we include and what should we

include?

Jeff Knight, Adam Scaife and Chris Folland

Hadley Centre for Climate Prediction and Research

2

Outline

Two sets of Hadley Centre C20C runs: ‘natural’ and ‘all’ forcings.

Natural forcings Anthropogenic forcings Sea surface momentum exchange Other possible forcings and future plans

3

Natural Forcings

Sea surface Temperature (SST) Sea-Ice anomalies Milankovitch cycles Total solar irradiance Stratospheric volcanic aerosols

4

SST and Sea-Ice forcings

HadISST1.1 (Rayner et al., 2003) 1870-present.

Reduced space optimal interpolation (RSOI) for filling data

gaps.

Karl Taylor variance correction method applied to model

input.

Forcing specified monthly, interpolated daily by the model.

Really a mixture of natural and anthropogenic forcings but

treated as natural here.

5

Milankovitch cyclesObliquity, eccentricity and date of perihelion are all included.

Affects seasonal/latitudinal distribution of solar radiationbut not the annual mean.

Mostly luni-solar precessional effect on time of perihelion: 25 minutes per year.

Accurate.

6

Milankovitch ForcingTop of atmosphere forcing changes wrt 1975

0.6 Wm-2 warming over NHtropics in MAM 1870-2000.

To 0.9 Wm-2 cooling over Arctic in JJA.

Similar warming in SH in Nov and Dec.

Small but easy and included for regional detail?

7

Solar ForcingReconstruction updated from Lean et al. (1995) to 1998, constant after.

11-year cycle reconstruction based on astronomical sunspot observations and a calibration based on modern satellite data.

Long-term change in line with activity indicated by number of sunspot groups and assuming 3.3 Wm-2 difference for Maunder Minimum based on sun-like stars.

Increase shortest wavebands more, via relationship from recent solar cycles. But limited stratospheric resolution and no interactive ozone.

Increases by 0.3 Wm-2 1900-1950. Relatively small but rather uncertain. Other reconstructions (e. g. Hoyt & Schatten, 1993) differ.

8

Volcanic ForcingAerosol optical depth at 550 nm(Crowley, 2000). Based on polar ice-core sulphate to 1960, ground based and satellite radiance after (Sato et al., 1993).

Monthly, averaged to 0°-30°, 30°-90° bands used in radiation scheme.

Leads to a variable negative surface climate forcing due to shortwave reflection.

Moderate accuracy.

Other series (e.g. Robock & Free, 1996) weight eruptions differently.

9

Natural radiative forcing

Instantaneous annual mean total (SW+LW) radiative forcing wrt 1949 (Wm-2)

(No stratospheric adjustment)

Diagnosed from double radiation call in 1 ensemble member

10

Anthropogenic Forcings

Greenhouse gases Tropospheric and stratospheric ozone Direct and indirect sulphate aerosols Land surface characteristics

11

Greenhouse gasesChanges in CO2, CH4, N2O, CFC11, CFC12, CFC113, HCFC22, HFC125 and HFC134A are all included as in Johns et al., (2003).

CO2 : 284.5 to 364.7 ppmv (1859-2000)

CH4 : 0.486 to 0.966 ppmm

N2O : 425.7 to 483.2 ppbm

CFC11 & 12 increase from 1950 and decline after 1998.

CFC113 increases from 1980 and declines after 2000.

HCFCs increase from the late 20th century.

Well known.

12

Ozone Forcing% changes shown wrt 1970

STOCHEM simulation for troposphere.

Linear trends from Randel et al. (1998) for stratosphere after 1975, half trends 1970-75. Combination of SAGE and ozonesonde profiles plus TOMS total ozone.

Stratospheric ozone loss plus tropospheric increase.

Recently discovered error in converting Randel data gives 1.8x forcing in stratosphere!!

Moderately well known.

13

Direct aerosol forcingRadiative impact of anthropogenic sulphate aerosols.

Surface and high-level (~930 hPa from 1975) SO2 emissions history as in Johns et al. (2003).

STOCHEM derived seasonal climatology of oxidants used to generate H2SO4.

Full aerosol scheme treating the formation, transportand fate of aerosols.

Moderately accurate.

14

Indirect aerosol forcingEffect of aerosol on CCN and hence changes to cloud albedo (first indirect effect).

Applied as a 3D field of cloud effective albedo changes computed off-line.

Begins in N mid-latitudes.

Concentrated over continentsand mainly affects low-midlevel cloud.

Less accurate.

15

Surface forcings Due to land use changes e.g. deforestation Based on satellite measurements and current

and historical records of land use Changing surface scheme parameters:

Vegetation fraction Root Depth Leaf area index

Infiltration factor Surface Capacity Evaporative resistance

Snow-free albedo Deep snow albedo

Roughness length Canopy height

16

Surface albedo forcingChanges wrt 1970.

Deforestation leads to increased albedo in snowy conditions.

Defined globally but onlyimpacts snow covered regions.

Shows N. American andE. European clearances in 19th century.

17

Surface roughness changes

Changes wrt 1970

Deforestation leads to shorterroughness lengths and so reduced atmospheric friction.

18

‘All forcings’ radiative forcingInstantaneous annual mean total (SW+LW) radiative forcing wrt 1949 (Wm-2) for ‘all forcings’ case.

Influence of increasing GHGs

Volcanoes

Aerosol decline near 45°N

19

Surface roughness change over oceans

Previous studies suggest that the ocean-atmosphere coupling may be tooweak in the HadAM3 version of the unified model (e.g. Rodwell and Folland, 2002).

We therefore increase the coupling between ocean and atmosphere bydoubling the Charnock parameter in all runs:

Zmsea = max ( Zmmin , Cτ/gρ*)

Where Z = surface roughness lengthC = Charnock parameter ~ 0.02

τ = surface wind stressg = acc’n due to gravity ρ* = surface density

20

Other forcings 2nd indirect effect of sulphate aerosols – reduced

conversion efficiency to precipitable drop size leading to higher liquid water contents, brighter and longer-lived clouds

Aerosol effects on ice clouds? Almost completely unknown DMS emissions, other aerosols including black carbon,

biomass burning aerosol, sea salt, dust Solar/cosmic high energy particles? Prescribed land surface anomalies

Comprehensive approach vs. uncertainty

21

Black Carbon Forcing

Absorbing aerosol so positive climate forcing offsetting sulphates

Some very large regional burdens

‘Asian Brown Cloud’

22

HadGEM1 Hadley Centre Global Environmental Model v.1 Atmospheric component HadGAM1 New dynamical core, resolution, improved

parameterisations, new carbon cycle and chemistry options, etc i.e. very different to HadAM3.

Allows direct and indirect aerosol effects to be computed directly by the model sulphur scheme.

Can also include black carbon, biomass burning aerosol, sea salt and dust.

To be used in next phase of C20C.

23

Conclusions 6 members 1950-2002 and 6 1870-2002 now complete

for ‘all forcings’, 6 members 1950-2002 ‘natural’. Almost all known natural forcings of climate are included. The most important anthropogenic effects have been

included in the ‘all-forcings’ ensemble. A thorough treatment of land use changes is included for

the first time in transient runs. Various levels of uncertainty in the forcings. Additional uncertainty from forcings which are not

included. Favour continuing comprehensive approach to reflect the

state of the science and to enhance regional studies.