2 2 2 2 (ppmv) (ppmv)CO (ppmv)CO CO (ppmv)CO

Simulated changes in vegetation distribution, land carbon storage, and atmospheric CO ₂ in response to a collapse

of the North Atlantic thermohaline circulation

P. K¨ ohler

, F. Joos

, S. Gerber

& R. Knutti

1: Alfred Wegener Institute for Polar and Marine Research, P.O. Box 12 01 61, D-27515 Bremerhaven, Germany, email: pkoehler@awi-bremerhaven.de 2: Climate and Environmental Physics, Physics Institute, University of Bern, Sidlerstr. 5, 3012 Bern, Switzerland

Measurements on air enclosed in glacial ice show that atmospheric CO2varied by 20 ppmv within several millenia with large iceberg discharges into the North Atlantic (NA) during Heinrich events 4 to 6. The iceberg discharges have been linked to changes in the NA Thermohaline Circulation (THC). Here, we analyse how abrupt changes in the NA THC affect the terrestrial carbon cycle by forcing the Lund- Potsdam-Jena Dynamic Global Vegetation Model with climate perturbations from freshwater experiments with the ECBILT-CLIO ocean-atmosphere model. Changes in the marine carbon cycle are not addressed. Modelled NA THC collapsed and recovered after about a millennium in response to prescribed freshwater forcing pre- turbing glacial background climate. The initial cooling of several Kelvin over Eurasia causes a reduction of extant boreal and temperate forests and a decrease in carbon storage in high northern latitudes, whereas improved growing conditions and slower soil decomposition rates lead to enhanced storage in mid-latitudes. The magnitude and evolution of global terrestrial carbon storage in response to abrupt THC changes depends sensitively on the initial climate conditions which are here varied between preindustrial and glacial background climate. Terrestrial storage varies between 67 and +50 PgC for a range of experiments that start at different times during the last 21,000 years. Simulated peak-to-peak differences in atmospheric CO2andδ¹³C are between 6 and 18 ppmv and 0.18 and 0.30hand compatible with the ice core records.

Forcing

-42 -40 -38 -36

δ18Ο (ο/οο)

-42 -41 -40 -39 -38

δ18Ο (ο/οο)

60 50 40 30 20

GISP2 age (kyr BP) 190

200 210 220

CO2 (ppmv)

A4 A3 A2 A1

▲ ▲ ▲ ▲ ▲

12 8

17 14

H6 H5a▼ H5 H4 H3 H2

GISP2 - Northern T

Byrd - Southern T

Taylor Dome - CO₂ 180 200 220 240 260 280 300

CO2 (ppmv)

180 200 220 240 260 280 300

-10 -5 0 5 10 15 20

∆Area (1012 m2)

-12 -10 -8 -6 -4 -2 0

∆T (K)

20 15 10 5

Time (kyr BP) -12

-10 -8 -6 -4 -2 0

global land 30^oS-30^oN 30^oN-90^oN

20 15 10 5 0

Time (kyr BP) -175 -150 -125 -100 -75 -50 -25 0

∆prec (mm yr-1)

A B

C D

ice retreat

net change

sea level

Figure 1: Top Left: (Motivation of this study) Ice core records of fast changes in northern and southern temperature and CO2. D/O events, Heinrich events and Antarctic warm events are labelled.

Top right: Different initial conditions for last 21 kyr. LPJ was forced by anomalies calculated with the HadSM3 climate model for conditions during the last 21 kyr BP.

Bottom: Temperature and precipitation (not shown) anomalies from ECBILT-CLIO were used as climate anomalies in LPJ-DGVM.

Glacial – Interglacial

20000 15000 10000 5000 0

Time (yr BP) 2000

2100 2200 2300 2400 2500 2600 2700 2800 2900 3000

Terrestrial carbon (PgC)

Figure 2: Top: Glacial/interglacial carbon storage during the last 21 kyr (blue).

Carbon storage anomalies during collapse of THC at different initial conditions (red, black, cyan, magenta). Conditions (1, 13, 17, 21 kyr BP) in red were analysed in detail (see CO2 in Figure 4). Middle, bottom: Terrestial carbon storage at 1 and 21 kyr BP, respectively.

Preindustrial reference

-60 -40 -20 0 20 40 60 80

Latitude -0,4

-0,3 -0,2 -0,1 0 0,1 0,2

Anomaly in relative foliar projective cover

tropical trees temperate trees

grasses

boreal trees 1 kyr BP

-80 -60 -40 -20 0 20 40 60 80

Anomaly (PgC)

0 1000 2000 3000

Simulation time (yr) -80

-60 -40 -20 0 20 40 60 80

total soil+litter vegetation

1 kyr BP

Figure 3: Top left: Changes in vegetation cover for aggregated PFTs.

Top right: Carbon storage anomaly (total, soil, vegetation).

Middle: Relative change in tree cover (peak THC shutdown compared with initial cond.).

Bottom: Carbon storage anomaly over time. Principal response is similar for dif- ferent initial conditions, but main changes in the northern hemisphere are due to land ice sheets shifting further south during glacial conditions.

Atmospheric CO₂

275 280 285 290 295 300

CO2 (ppmv)

CO₂ fert. feedbacks off CO2 fert feedbacks on

230 235 240 245 250

CO2 (ppmv)

195 200 205 210 215

CO2 (ppmv)

0 1000 2000 3000 Simulation time (yr) 175

180 185 190 195 200

CO2 (ppmv)

1 kyr BP

13 kyr BP

21 kyr BP 17 kyr BP

Figure 4: Final results for changes in atmospheric CO2 induced through changes in terrestrial carbon storage for different initial conditions.