Generated using version 3.0 of the official AMS L A T E X template Intensification of North American megadroughts through surface and dust aerosol forcing. Benjamin I Cook * NASA Goddard Institute for Space Studies, NY, NY, USA Lamont-Doherty Earth Observatory, Palisades, NY, USA Richard Seager Lamont-Doherty Earth Observatory, Palisades, NY, USA Ron L Miller NASA Goddard Institute for Space Studies, NY, NY, USA Joseph A Mason University of Wisconsin, Madison, WI, USA * Corresponding author address: Benjamin I Cook, NASA Goddard Institute for Space Studies, 2880 Broadway, New York, NY 10025. E-mail: [email protected]1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Generated using version 3.0 of the official AMS LATEX template
Intensification of North American megadroughts through surface
and dust aerosol forcing.
Benjamin I Cook ∗
NASA Goddard Institute for Space Studies, NY, NY, USA
Lamont-Doherty Earth Observatory, Palisades, NY, USA
Richard Seager
Lamont-Doherty Earth Observatory, Palisades, NY, USA
Ron L Miller
NASA Goddard Institute for Space Studies, NY, NY, USA
Joseph A Mason
University of Wisconsin, Madison, WI, USA
∗Corresponding author address: Benjamin I Cook, NASA Goddard Institute for Space Studies, 2880
Palmer, W., 1965: Meteorological drought, Research Paper No. 45. US Weather Bureau,
Washington, DC, 58.
Ross, T. and N. Lott, 2003: A climatology of 1980-2003 extreme weather and climate events.
Tech. rep., NOAA/NESDIS. National Climatic Data Center, Asheville, NC.
Ruiz-Barradas, A. and S. Nigam, 2006: IPCC’s Twentieth-Century Climate Simulations:
26
Varied Representations of North American Hydroclimate Variability. Journal of Climate,
19 (16), 4041–4058.
Schmidt, G., et al., 2006: Present-day atmospheric simulations using GISS ModelE: Com-
parison to in situ, satellite, and reanalysis data. Journal of Climate, 19 (2), 153–192.
Schubert, S., M. Suarez, P. Pegion, R. Koster, and J. Bacmeister, 2004: On the Cause of
the 1930s Dust Bowl. Science, 303 (5665), 1855–1859.
Schubert, S., et al., 2009: A US CLIVAR Project to Assess and Compare the Responses of
Global Climate Models to Drought-Related SST Forcing Patterns: Overview and Results.
Journal of Climate, 22 (19), 5251–5272.
Seager, R., R. Burgman, Y. Kushnir, A. Clement, E. Cook, N. Naik, and J. Miller, 2008a:
Tropical Pacific forcing of North American medieval megadroughts: Testing the con-
cept with an atmosphere model forced by coral-reconstructed SSTs. Journal of Climate,
21 (23), 6175–6190.
Seager, R., Y. Kushnir, C. Herweijer, N. Naik, and J. Velez, 2005: Modeling of Tropical
Forcing of Persistent Droughts and Pluvials over Western North America: 1856–2000.
Journal of Climate, 18 (19), 4065–4088.
Seager, R., Y. Kushnir, M. Ting, M. Cane, N. Naik, and J. Miller, 2008b: Would Advance
Knowledge of 1930s SSTs Have Allowed Prediction of the Dust Bowl Drought? Journal
of Climate, 21 (13), 3261–3281.
Seager, R., et al., 2007: Model projections of an imminent transition to a more arid climate
27
in southwestern North America. Science, 316 (DOI: 10.1126/science.1139601), 1181–
1184.
Sophocleous, M., 2010: Review: groundwater management practices, challenges, and inno-
vations in the High Plains aquifer, USA—lessons and recommended actions. Hydrogeology
Journal, 18 (3), 559–575.
Stahle, D., F. Fye, E. Cook, and R. Griffin, 2007: Tree-ring reconstructed megadroughts
over north america since ad 1300. Climatic change, 83 (1), 133–149.
Thornthwaite, C., 1948: An approach toward a rational classification of climate. Geographical
review, 38 (1), 55–94.
Van Auken, O., 2000: Shrub invasions of north american semiarid grasslands. Annual Review
of Ecology and Systematics, 31, 197–215.
Wisser, D., B. Fekete, C. Vorosmarty, and A. Schumann, 2010: Reconstructing 20th century
global hydrography: a contribution to the Global Terrestrial Network-Hydrology (GTN-
H). Hydrol. Earth Syst. Sci, 14, 1–24.
28
List of Tables
1 Summary of model experiments conducted as part of this study.
All experiments represent 5-member ensembles, with each ensem-
ble initialized using a unique set of starting conditions. For each
simulation, the table indicates the years over which the ensembles
were run, the source of the SST forcing, and if bare soil (BSOIL)
or dust source (DUST) boundary conditions were included. 30
29
Table 1. Summary of model experiments conducted as part of this study. Allexperiments represent 5-member ensembles, with each ensemble initialized usinga unique set of starting conditions. For each simulation, the table indicates theyears over which the ensembles were run, the source of the SST forcing, and ifbare soil (BSOIL) or dust source (DUST) boundary conditions were included.Experiment Years SSTs Dune,
Bare SoilDune,Dust
SST-MOD 1857-2005 Kaplan et al. (1998) – –
SST-Only 1321-1461 Cobb et al. (2003) – –
SST+BSOIL 1321-1461 Cobb et al. (2003) + –
SST+BSOIL+DUST 1321-1461 Cobb et al. (2003) + +
30
List of Figures
1 NINO 3.4 sea surface temperature (SST) anomalies (a) and dune mobiliza-
tion region (b) for the MCA experiments. All MCA experiments (SST-Only,
SST+BSOIL, SST+BSOIL+DUST) use the same SST forcing, based on a
coral reconstruction of SSTs in the tropical Pacific. Anomalies are relative
to the 1856-2005 mean SSTs in the NINO3.4 region from the Kaplan SST
dataset. Dune mobilization region is used to define the 50% bare soil area in
the SST+BSOIL and SST+BSOIL+DUST experiments and the dust source
region in the SST+BSOIL+DUST experiment. 35
2 Monthly temperature (a) and precipitation (b) climatology (1961-1990) for the
Central Plains region (105oW-95oW, 32oN-44oN), calculated from our modern
day SST forced ensemble and the CRU 2.1 climate grids (green line). 36
3 Summer season (June-July-August, JJA) PDSI anomalies from the North
American Drought Atlas (NADA), averaged over the Central Plains region
(105oW-95oW, 32oN-44oN). Droughts in this region during the late Medieval
Climate Anomaly (MCA) (a) were generally more persistent than droughts
after 1500 C.E. (b). Both time series were smoothed with a 5-year low pass
(lowess spline) filter. 37
31
4 Autocorrelation function (ACF) for Central Plains PDSI from the NADA.
Droughts during the MCA (a) have significant persistence (grey line, p ≤ 0.05)
out to a lag of -9 years. Post-MCA drought persistence (b) is significantly
reduced, with significant persistence only at a lag of -1 year. Even after
standardizing to zero mean, the MCA PDSI (c) retains significant persistence
at lags of -1, -2, -3, and -8 years, while the post-MCA PDSI (d) is relatively
unchanged. 38
5 Ensemble mean monthly temperature (K) (a) and precipitation (mm day-1)
(b) response over the Central Plains region (105oW-95oW, 32oN-44oN), in our
SST-Only (blue bars), SST+BSOIL (green bars), and SST+BSOIL+DUST
(brown bars) experiments, relative to SST-MOD. Anomalies significant differ-
ent from SST-MOD (Student’s two-side t-test, p ≤ 0.05) are indicated with a
*. 39
6 Ensemble mean monthly surface temperature (JJA, K) (a,c,e) and precipita-
tion (June, mm day-1) (b,d,f) responses over North American in our MCA
model runs. The central Plains region (105oW-95oW, 32oN-44oN) is outlined
in the black dashes. Cells with insignificant (p ≥ 0.05) differences between
the MCA runs and SST-MOD have been masked out. 40
7 Ensemble mean time series of mean monthly surface temperature (JJA, K)
(a) and precipitation (June, mm day-1) (b) responses over North American
in our MCA model runs, averaged over the Central Plains region (105oW-
95oW, 32oN-44oN). All time series have been smoother with a 5-year low pass
(lowess) filter. 41
32
8 Normalized histograms of (a) mean monthly surface temperature (JJA, K), (b)
precipitation (June, mm day-1), (c) mean monthly maximum surface temper-
ature (JJA, K), and (d) mean monthly minimum surface temperature (JJA,
K) responses over North American in our MCA model runs, averaged over
the Central Plains region (105oW-95oW, 32oN-44oN). 42
9 Normalized histograms of JJA surface flux (W m-2) responses averaged over
the Central Plains region (105oW-95oW, 32oN-44oN): (a) latent heat flux,
(b) sensible heat flux, (c) incident shortwave flux at the surface, and (d)
downwelling longwave flux at the surface. 43
10 Normalized histograms of mean June surface climate responses averaged over
the Central Plains region (105oW-95oW, 32oN-44oN): (a) net radiation at the
surface (W m-2), (b) planetary boundary layer height (meters), (c) peak lapse
rate in the planetary boundary layer (K km-1), and (d) convective cloud cover
(% normal departure). 44
11 Ensemble average summer season (JJA) PDSI calculated from our MCA and
averaged over the Central Plains region (105oW-95oW, 32oN-44oN): (a) SST-
Only, (b) SST+BSOIL, and (c) SST+BSOIL+DUST. PDSI anomalies for all
MCA runs were normalized relative to 1857-2005 C.E. from the SST-MOD
ensemble. 45
33
12 Autocorrelation function (ACF) for the ensemble average JJA Central Plains
PDSI (105oW-95oW, 32oN-44oN) from the SST-Only (a), SST+BSOIL (b),
and SST+BSOIL+DUST (c) model runs. PDSI in the SST+BSOIL+DUST
ensemble, driven by the dust aerosol forced drying, is the only scenario that
can generate significant persistence at time scales similar to the MCA PDSI
from the NADA. Even after the adjustment to zero mean (d,e,f), SST+BSOIL+DUST
is the only the scenario with significant persistence at lags of -2 and -3 years,
similar to the autocorrelation of the adjusted NADA PDSI. 46
34
Model Boundary Conditions
144oW 126
oW 108
oW 90
oW 72
oW 54
oW
20oN
30oN
40oN
50oN
60oN
144oW 126
oW 108
oW 90
oW 72
oW 54
oW
20oN
30oN
40oN
50oN
60oN
Region of Dune Mobilization
1320 1340 1360 1380 1400 1420 1440 1460
−1
−0.5
0
0.5
1
Annual NINO 3.4 Anomalies (MCA)
K
(a)
(b)
Fig. 1. NINO 3.4 sea surface temperature (SST) anomalies (a) and dune mobilizationregion (b) for the MCA experiments. All MCA experiments (SST-Only, SST+BSOIL,SST+BSOIL+DUST) use the same SST forcing, based on a coral reconstruction of SSTsin the tropical Pacific. Anomalies are relative to the 1856-2005 mean SSTs in the NINO3.4region from the Kaplan SST dataset. Dune mobilization region is used to define the 50%bare soil area in the SST+BSOIL and SST+BSOIL+DUST experiments and the dust sourceregion in the SST+BSOIL+DUST experiment.
35
J F M A M J J A S O N D−5
0
5
10
15
20
25
30
oC
Surface Temperature
J F M A M J J A S O N D0.5
1
1.5
2
2.5
3
3.5
mm day−1
Precipitation
Model
CRU 2.1
Central Plains Climatology (1961-1990)
(a) (b)
Fig. 2. Monthly temperature (a) and precipitation (b) climatology (1961-1990) for theCentral Plains region (105oW-95oW, 32oN-44oN), calculated from our modern day SST forcedensemble and the CRU 2.1 climate grids (green line).
36
Central Plains PDSI (NADA, JJA)
MCA (1100-1500 C.E.)
Post-MCA (1501-2000 C.E.)
1550 1600 1650 1700 1750 1800 1850 1900 1950 2000
−3
−2
−1
0
1
2
3
1100 1150 1200 1250 1300 1350 1400 1450 1500
−3
−2
−1
0
1
2
3
(a)
(b)
Fig. 3. Summer season (June-July-August, JJA) PDSI anomalies from the North AmericanDrought Atlas (NADA), averaged over the Central Plains region (105oW-95oW, 32oN-44oN).Droughts in this region during the late Medieval Climate Anomaly (MCA) (a) were generallymore persistent than droughts after 1500 C.E. (b). Both time series were smoothed with a5-year low pass (lowess spline) filter.
37
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
Autocorrelation Function (NADA PDSI) MCA (1100−1500 C.E.) Post-MCA (1501−2000 C.E.)
(a) (b)
Lag
Default Mean
Zero Mean
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0(c) (d)
Fig. 4. Autocorrelation function (ACF) for Central Plains PDSI from the NADA. Droughtsduring the MCA (a) have significant persistence (grey line, p ≤ 0.05) out to a lag of -9 years.Post-MCA drought persistence (b) is significantly reduced, with significant persistence onlyat a lag of -1 year. Even after standardizing to zero mean, the MCA PDSI (c) retainssignificant persistence at lags of -1, -2, -3, and -8 years, while the post-MCA PDSI (d) isrelatively unchanged.
38
(a)
(b)
Central Plains ΔT and ΔP, vs SST-MOD
J F M A M J J A S O N D−0.4
−0.2
0
0.2
Precipitation
m
m d
ay
−1
SST−Only
SST+BSOIL
SST+BSOIL+DUST
J F M A M J J A S O N D−0.4
0
0.4
0.8
1.2
Surface Temperature
K
* * * ** *
*
** *
**
*
*
*
**
*
*
*
** *
*
*
* *
* *
**
*
*
*
*
*
* *
**
*
* *
Fig. 5. Ensemble mean monthly temperature (K) (a) and precipitation (mm day-1) (b)response over the Central Plains region (105oW-95oW, 32oN-44oN), in our SST-Only (bluebars), SST+BSOIL (green bars), and SST+BSOIL+DUST (brown bars) experiments, rela-tive to SST-MOD. Anomalies significant different from SST-MOD (Student’s two-side t-test,p ≤ 0.05) are indicated with a *.
Fig. 6. Ensemble mean monthly surface temperature (JJA, K) (a,c,e) and precipitation(June, mm day-1) (b,d,f) responses over North American in our MCA model runs. Thecentral Plains region (105oW-95oW, 32oN-44oN) is outlined in the black dashes. Cells withinsignificant (p ≥ 0.05) differences between the MCA runs and SST-MOD have been maskedout.
40
Central Plains Climate Anomalies
1321 1341 1361 1381 1401 1421 1441 1461−0.4
0
0.4
0.8
1.2
1.6
2
2.4
Surface Temperature (JJA)
K
1321 1341 1361 1381 1401 1421 1441 1461
−0.8
−0.6
−0.4
−0.2
0
0.2
0.4
Precipitation (June)
m
m d
ay
−1
SST-Only
SST+BSOIL
SST+BSOIL+DUST
(a)
(b)
Fig. 7. Ensemble mean time series of mean monthly surface temperature (JJA, K) (a) andprecipitation (June, mm day-1) (b) responses over North American in our MCA model runs,averaged over the Central Plains region (105oW-95oW, 32oN-44oN). All time series have beensmoother with a 5-year low pass (lowess) filter.
41
Temperature and Precipitation Anomalies
−2 −1 0 1 2 3 40
0.05
0.1
0.15
0.2
Temperature, Mean (JJA)
R
el F
req
K
−2 −1 0 1 20
0.05
0.1
0.15
0.2
0.25
Precipitation (June)
R
el F
req
mm day−1
−2 −1 0 1 2 3 40
0.05
0.1
0.15
0.2
Temperature, Max (JJA)
R
el F
req
K−2 −1 0 1 2 3 4
0
0.05
0.1
0.15
0.2
0.25
Temperature, Minimum (JJA)
R
el F
req
K
SST-Only
SST+BSOIL
SST+BSOIL+DUST
(a) (b)
(d)(c)
Fig. 8. Normalized histograms of (a) mean monthly surface temperature (JJA, K), (b)precipitation (June, mm day-1), (c) mean monthly maximum surface temperature (JJA,K), and (d) mean monthly minimum surface temperature (JJA, K) responses over NorthAmerican in our MCA model runs, averaged over the Central Plains region (105oW-95oW,32oN-44oN).
42
Surface Fluxes (JJA)
−40 −20 0 20 400
0.05
0.1
0.15
0.2
Latent Heat Flux
R
el F
req
W m−2
−30 −20 −10 0 10 20 300
0.05
0.1
0.15
0.2
0.25
0.3
Sensible Heat Flux
R
el F
req
W m−2
−30 −20 −10 0 10 20 300
0.05
0.1
0.15
0.2
Incident Shortwave, Surface
R
el F
req
W m−2
−20 −10 0 10 200
0.05
0.1
0.15
0.2
0.25
0.3
Downwelling Longwave, Surface
R
el F
req
W m−2
SST-Only
SST+BSOIL
SST+BSOIL+DUST
(a) (b)
(d)(c)
Fig. 9. Normalized histograms of JJA surface flux (W m-2) responses averaged over theCentral Plains region (105oW-95oW, 32oN-44oN): (a) latent heat flux, (b) sensible heat flux,(c) incident shortwave flux at the surface, and (d) downwelling longwave flux at the surface.
43
Surface Response (June)
−100 0 100 200 3000
0.05
0.1
0.15
0.2
PBL Height
R
el F
req
meters
−2 −1 0 10
0.1
0.2
0.3
0.4
Peak Lapse Rate, PBL
R
el F
req
K km−1
−75 −50 −25 0 25 50 750
0.05
0.1
0.15
0.2
Convective Cloud Cover
R
el F
req
% normal
−30 −20 −10 0 100
0.05
0.1
0.15
0.2
0.25
Net Radiation, Surface
R
el F
req
W m−2
SST-Only
SST+BSOIL
SST+BSOIL+DUST
(a) (b)
(d)(c)
Fig. 10. Normalized histograms of mean June surface climate responses averaged over theCentral Plains region (105oW-95oW, 32oN-44oN): (a) net radiation at the surface (W m-2),(b) planetary boundary layer height (meters), (c) peak lapse rate in the planetary boundarylayer (K km-1), and (d) convective cloud cover (% normal departure).
44
1340 1360 1380 1400 1420 1440 1460
−3
−2
−1
0
1
2
SST+BSOIL+DUST
Central Plains PDSI (JJA, MCA Model Runs)
1340 1360 1380 1400 1420 1440 1460
−3
−2
−1
0
1
2
SST+BSOIL
1340 1360 1380 1400 1420 1440 1460
−3
−2
−1
0
1
2
SST-Only
(a)
(b)
(c)
Fig. 11. Ensemble average summer season (JJA) PDSI calculated from our MCA andaveraged over the Central Plains region (105oW-95oW, 32oN-44oN): (a) SST-Only, (b)SST+BSOIL, and (c) SST+BSOIL+DUST. PDSI anomalies for all MCA runs were nor-malized relative to 1857-2005 C.E. from the SST-MOD ensemble.
45
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
Autocorrelation Function, Model PDSI (Ensemble Mean)
SST-Only SST+BSOIL SST+BSOIL+DUST
(a) (b) (c)
Lag
Default Mean
Zero Mean
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7 8 9 10−0.2
0
0.2
0.4
0.6
0.8
1.0(d) (e) (f)
Fig. 12. Autocorrelation function (ACF) for the ensemble average JJA CentralPlains PDSI (105oW-95oW, 32oN-44oN) from the SST-Only (a), SST+BSOIL (b), andSST+BSOIL+DUST (c) model runs. PDSI in the SST+BSOIL+DUST ensemble, driven bythe dust aerosol forced drying, is the only scenario that can generate significant persistenceat time scales similar to the MCA PDSI from the NADA. Even after the adjustment to zeromean (d,e,f), SST+BSOIL+DUST is the only the scenario with significant persistence atlags of -2 and -3 years, similar to the autocorrelation of the adjusted NADA PDSI.