Climatic Consequences of Gradual Conversion of Amazonian Tropical Forests Into Degraded Pasture or...

26
Climatic Consequences of Gradual Conversion of Amazonian Tropical Forests Into Degraded Pasture or Soybean Cropland: a new Vegetation-Climate Equilibrium State in Amazonia Gilvan Sampaio ([email protected]) Carlos Nobre Prakki Satyamurty Instituto Nacional de Pesquisas Espaciais - Instituto Nacional de Pesquisas Espaciais - INPE INPE Centro de Ciência do Sistema Terrestre - Centro de Ciência do Sistema Terrestre - CCST CCST Conferência Científica Internacional – LBA-GEOMA-PPBio Manaus-AM / Novembro de 2008

Transcript of Climatic Consequences of Gradual Conversion of Amazonian Tropical Forests Into Degraded Pasture or...

Climatic Consequences of Gradual Conversion of Amazonian Tropical Forests Into Degraded Pasture or Soybean Cropland: a new Vegetation-Climate

Equilibrium State in Amazonia

Gilvan Sampaio([email protected])

Carlos Nobre

Prakki Satyamurty

Instituto Nacional de Pesquisas Espaciais - INPEInstituto Nacional de Pesquisas Espaciais - INPECentro de Ciência do Sistema Terrestre - CCSTCentro de Ciência do Sistema Terrestre - CCST

Conferência Científica Internacional – LBA-GEOMA-PPBio

Manaus-AM / Novembro de 2008

Oyama and Nobre, 2003

Two Biome-Climate Equilibrium States found for South America!

b) ‘Savannization’ of Amazonia and ‘desertification’ in NE Brazil

(b) Second State - Biome-climate equilibrium starting from desert land cover as

Initial Condition for the Dynamic Vegetation Model

(a) First State - Biome-climate equilibrium starting from forest land cover as initial condition for the Dynamic Vegetation Model. These results are similiar to current natural

vegetation.

Question: What are the conditions for deforestation to induce an ‘abrupt’ transition to the second biome-climate stable equilibrium?

?

The study is divided in two parts:

(I)the numerical simulations kept the biomes unchanged throughout the integration and did not interact with climate. We employed the CPTEC-INPE AGCM to assess the effects of Amazonian deforestation on the regional and global climate, using two ways to assign land cover: 1) land cover projections of deforestation and replacement by degraded pasture or soybean cropland from a business-as-usual scenario of future deforestation and 2) land cover projections from random scenarios of deforestation and replacement by degraded pasture.

(II)the biomes were allowed to interact with climate freely, therefore they can change during the integration. To do this, the CPTEC-PVM was asynchronously coupled to the CPTEC AGCM.

PART I

PROJECTED DEFORESTATION SCENARIOS

We used land cover change scenarios with deforested areas equal to 20%, 40%, 50%, 60%, 80% and 100% of the original extent of the Amazon forest.

The land cover change scenarios with deforested areas smaller than 40% are from Soares-Filho et al. [2006].

The land cover change scenarios with deforested areas greater than 40% are obtained using the same methodology extending further into the future the simulation of deforestation under the business-as-usual scenario, which considers that recent deforestation trends will continue; highways currently scheduled for paving will be paved; compliance with legislation requiring forest reserves on private land will remain low; and protected areas will not be enforced Sampaio et al., 2007

Geophys. Res. Lett., 34

PROJECTED DEFORESTATION SCENARIOS

Pasture or Soybean

Evergreen Broadleaf Trees

Sampaio et al., 2007Geophys. Res. Lett., 34

Random Deforestation Scenarios (example of 50% deforestation)

Sampaio 2008 PhD Thesis

Global Circulation Model – CPTEC/INPE 1.0 (Kinter et al. 1997, Bonatti, 1996, Cavalcanti et al., 2002)

• Model horizontal resolution: T062 (~200km x 200 km)• Vertical levels: 42• Land surface scheme: Simplified Simple Biosphere Model (SSiB) – Xue et

al., 1991

• For each land grid point, a vegetation type (biome) is prescribed, and the vegetation classification follows Dorman and Sellers (1989). A set of physical, morphological, and physiological parameters is assigned to each biome.

• Boundary conditions:Monthly SST: climatological - NCEP-Reynolds e Smith, 1994Initial climatological values: soil moisture, albedo and snow depth.Sea ice: considered at grid points for which SST is below -2ºC.

Sampaio et al., 2007Geophys. Res. Lett., 34

Part I - Experiments:

The CPTEC model was integrated for each experiment and control for 87 months with five different initial conditions derived from five consecutive days of NCEP analyses. The results are the mean of the last 60 months (experiment – control).

1) Control case: Today land cover scenario.

2) PASTURE: Land cover change scenarios with deforested areas equal to 20%, 40%, 50%, 60%, 80% and 100% of the original extent of the Amazon forest.

3) SOYBEAN: Land cover change scenarios with deforested areas of 20%, 50%, 80% and 100%.

For each scenario a pseudo-equilibrium between the climate and vegetation was obtained.

Sampaio et al., 2007Geophys. Res. Lett., 34

Season All Pasture All Soybean

JJA -27.5% -39.8%

SON -28.1% -39.9%

PrecipitationAmazonia - PASTUREArea: East/Northeast

0.6

0.7

0.8

0.9

1.0

1.1

1.2

0% 20% 40% 60% 80% 100%

Deforestation Area (%)

Rel

ativ

e P

reci

pit

atio

n (

p/p

0)

DJF

MAM

JJA

SON

Amazonia - SOYBEANArea: East/Northeast

0.6

0.7

0.8

0.9

1.0

1.1

1.2

0% 20% 40% 60% 80% 100%

Deforestation Area (%)

Rel

ativ

e P

reci

pit

atio

n (

p/p

0)

DJF

MAM

JJA

SON

PASTURE SOYBEAN

Relative Precipitation

The reduction in precipitation occurs mainly during the dry season, and is more evident when the deforested area is larger than 40% !

Sampaio et al., 2007Geophys. Res. Lett., 34

Season All Pasture All Soybean

JJA -15.7% -24.0%all Amazonia

ALEAT

Period All Pasture All Soybean

Annual -18,2% -25,8%

ASO -42,0% -47,4%

Precipitation (100% de desf.)

decrease in precipitation associated with pasture or

soybean expansion

PASTURE

PASTURE

SOYBEAN

Sampaio 2008 PhD Thesis

Searching for Multiple

Biome-Climate Equilibria

Climate from Degraded Pasture“Realistic” Deforestation Scenarios

Sampaio 2008 PhD Thesis

Post-Deforestation Equilibrium Potential Biomes

Climate from Degraded PastureRandom Deforestation Scenarios

Sampaio 2008 PhD Thesis

Post-Deforestation Equilibrium Potential Biomes

Climate from Soybean Crop“Realistic” Land Use Scenarios

Sampaio 2008 PhD Thesis

Post-Deforestation Equilibrium Potential Biomes

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Taxa de desflorestamento

Áre

a (

106 k

m2 )

Floresta

Savana

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Taxa de desflorestamento

Áre

a (1

06 k

m2)

Floresta

Savana

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Taxa de desflorestamento

Áre

a (1

06 k

m2)

Floresta

Savana

Climate from Degraded Pasture“Realistic” Deforestation Scenarios

Climate from Degraded PastureRandom Deforestation Scenarios

Climate from Soybean Crop“Realistic” Land Use Scenarios

Sampaio 2008 PhD Thesis

Area of Forest

Area of Savanna

Post-Deforestation Equilibrium Potential Biomes

PART II

Vegetation = f1 (climate variables) = f1 (g0, g5, Tc, h, s)

g0 = degree-days above 0 Cg5 = degree-days above 5 CTc = mean temperature of the coldest monthh = aridity index s = sesonality index

f1 is a highly nonlinear function

Climate = f2 (vegetation) = f2 (AGCM coupled to vegetated land surface scheme)

f2 is also a nonlinear function

CPTEC-PVM was coupled to the CPTEC AGCM

Vegetation ModelCPTEC PVM

AtmosphericModel

CPTEC AGCM

COUPLING

CPTEC-PVM was coupled to the CPTEC AGCM

-Control run: the present-day potential biomes (output of the PVM forced by the present-day climate) are kept unchanged during 20-year: five integrations with different initial conditions derived from five consecutive days of NCEP analyses.

-Fourteen 24-year integrations using dynamic vegetation; in these integrations, the biome distribution is updated every 6 years. Four iterations are found to be enough to reach a biosphere-atmosphere equilibrium state. The biome distribution and the climate of the last iteration are analyzed.

Perturbation Experiments:

CoupledPVM-AGCM Simulations

Initial Conditions Equlibrium State

Potential Biomes

Sampaio 2008 PhD Thesis

100% amzdes 80% amzdes 60% amzdes

50% amzdes 40% amzdes 20% amzdes

0% amzdes

Perturbation Experiments:Coupled PVM-AGCM Simulations

Sampaio 2008 PhD Thesis

100% amzflo 80% amzflo 60% amzflo

50% amzflo 40% amzflo 20% amzflo

0% amzflo

Perturbation Experiments:Coupled PVM-AGCM Simulations

Sampaio 2008 PhD Thesis

Sam

paio

200

8 P

hD T

hesi

s

These results suggest that the total deforested area of 50% is a sensitive point where there could be a change of states.

50%

50%

Tropical Forest Area

Rem

ain

ing

Tro

pic

al F

ore

st A

rea

(1

06 k

m2)

% of Altered Area in the Initial Condition

Sam

paio

200

8 P

hD T

hesi

s

CONCLUSIONS

The main impacts on Amazonia climate, because of deforestation, occur over eastern and central Amazonia, and are more evident when total deforested area is larger than 40%.

The results for eastern Amazonia, where changes in land cover are expected to be larger during this century, show increase in near-surface air temperature, and decrease in evapotranspiration and precipitation, which occurs mainly during the dry season for both classes of land use conversions.

We found two biosphere-atmosphere equilibrium states for South America as in Oyama and Nobre (2003): 1) present-day potential biomes 2) a new vegetation-climate equilibrium where the eastern part of the Amazonia tropical forest are replaced by tropical savannas, and semi-desert and desert areas appear in Northeast of Brazil.

In this study, the results suggest that 50% of total deforested area in Amazonia is the threshold for transition from present day potential biomes to a new vegetation-climate equilibrium state in Amazonia.