• No results found

Relationship of U.S. Summer Droughts with SST and Soil Moisture: Distinguishing the Time Scale of Droughts

N/A
N/A
Protected

Academic year: 2022

Share "Relationship of U.S. Summer Droughts with SST and Soil Moisture: Distinguishing the Time Scale of Droughts"

Copied!
43
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Relationship of U.S. Summer Droughts with SST and Soil Moisture:

Distinguishing the Time Scale of Droughts

Renguang Wu

Center for Ocean-Land-Atmosphere Studies

CTB Joint Seminar Series

February 24, 2009, NCEP

Co-author: James L. Kinter III

1

(2)

Why distinguish the time scale?

. Difference in impacts

. Difference in preferred regions

. Difference in mechanisms/factors

2

(3)

Mo&Schemm 2008

Wet Spell:PDSI>=2 Dry Spell:PDSI<=-2 Ratio of the total

number of months (1900-2004)

1-5 months Eastern US

6-11 months

> 1 year

Central-western

Difference in preferred regions

3

(4)

Difference in mechanisms

Roles of soil moisture for persistence of droughts:

Entekhabi et al. 1992 Findell and Eltahir 1997 Pal and Eltahir 2001 Schubert et al. 2004 Seager et al. 2005

--- soil moisture-precipitation feedback ---

4

(5)

P > n-month mean > transform > SPI * 1-month mean spi01 ** 2-month mean transform spi02 *** 3-month mean from Pearson spi03 ****** 6-month mean III distribution spi06 ********* 9-month mean to normal spi09 ************ 12-month mean distribution spi12

***……….*** 24-month mean spi24

How to distinguish the time scale?

Standardized Precipitation Index (SPI): defined for different time scales

5

(6)

Drought indices

PDSI (Palmer 1965): based on the water balance between supply and demand

SPI (McKee et al. 1993): based on precipitation only In the present study,

SPI03: short-term (<= 3months)

SPI09: medium-term (6-12 months) SPI24: long-term (> 1year)

6 Both PDSI and SPI are used in the drought monitoring

(7)

QuickTime™ and a decompressor are needed to see this picture.

QuickTime™ and a decompressor are needed to see this picture.

QuickTime™ and a decompressor are needed to see this picture.

QuickTime™ and a decompressor are needed to see this picture.

7

PDSI SPI09

SPI03 SPI24

Drought indices September 2008

(8)

Questions

(1) Which region SST has the most significant relationship with the U.S. summer droughts in Great Plains/Southwest?

(2) Which region summer droughts are mostly influenced by remote SST forcing (ENSO, Tropical IO, North Atlantic)?

(3) What is the role of soil moisture in droughts?

(4) How is the long-term change in the relationship between droughts and SST forcing?

8 What are the relative roles of remote SST forcing and regional

soil moisture in the droughts at different time scales?

Focus on summer droughts in this talk

(9)

Datasets

a. Drought indices (PDSI and SPI):

344 US Climate Divisions, 1895-2007, NCDC b. Soil moisture (/Evaporation)

344 US Climate Divisions, CPC bucket model, 1932-2005 19 stations in Illinois, 1981-2004

Grid (1/8 degree), Variable Infiltration Capacity (VIC) model of University of Washington, 1950-2000

c. SST

ERSST Version 2, 1854-2005 Hadley Center SST1, 1870-2006

9 Looking for statistically significant (robust) relationship

(10)

PDSI vs SPI:

local correlation

10 SPI01

SPI03 SPI02

SPI12 SPI09 SPI06

SPI24

Short-term: higher

correlation in eastern part

Medium-term and long-term:

higher correlation in central- western part

(11)

PDSI vs SPI:

standard deviation

11

SPI01 SPI06

SPI02 SPI09

SPI12

SPI24 SPI03

PDSI/3 PDSI: larger std in

central-western part SPI: relatively

uniform

(12)

PDSI vs SPI:

pattern

12 EOF1:

Continental

PDSI: larger loading in central-western SPI: larger loading in eastern

(13)

PDSI vs SPI:

pattern

13 EOF2: East-

west contrast

A trend:

Drying in the central-western Wetting in the eastern

Similar patterns

(14)

Example: 1930s

14 SPI01

SPI02

SPI03

PDSI/3

SPI06

SPI09

SPI12

SPI24 JJA 1933-1939

(15)

Example: 1950s

15 SPI01

SPI02

SPI03

PDSI/3

SPI06

SPI09

SPI12

SPI24 JJA 1952-1956

(16)

Drought-SST relationship

Schubert et al. 2008

C20C run NSIPP model OBS SST

GP Precipitation (>= 6yr)

OBS MODEL

SST Correlation (Annual Mean)

16

?

? ?

?

?

(17)

Drought-SST relationship

DJF SST JJA SST

PDSI

SPI03

SPI24 SPI09

17 JJA

Great Plains

cor with SST

(18)

Drought-SST relationship

DJF SST JJA SST

PDSI

SPI03

SPI09

SPI24 18 JJA

Southwest

cor with SST

(19)

Drought-soil moisture relationship

Findell & Eltahir’97

QuickTime™ and a decompressor

are needed to see this picture.

QuickTime™ and a decompressor

are needed to see this picture.

Initial Soil Saturation vs P(Initial-August23)

Soil Saturation(June25) vs P(June25-August23)

19

cc=0.55

Soil saturation =

moisture content/porosity

(20)

Validation:

bucket model vs obs Illinois all-months

20

bucket model

observation

(21)

Comparison:

bucket model vs VIC GP JJA

21

bucket model

VIC

(22)

Drought-soil moisture relationship

22 Great Plains JJA

(23)

Drought-soil moisture relationship

23 Southwest JJA

(24)

QuickTime™ and a decompressor are needed to see this picture.

Piechota & Dracup’96

QuickTime™ and a decompressor

are needed to see this picture.

QuickTime™ and a decompressor

are needed to see this picture.

ENSO-drought relationship

PDSI Composite

24

El Nino La Nina

(25)

Mo&Schemm 2008

ENSO-drought relationship

PDSI Composite:

Warm-Cold ENSO

OND

AMJ JFM

JJA

JAS

ASO

25 JFM vs JJA

(26)

ENSO-drought relationship

DJF NINO3.4 SST JJA NINO3.4 SST

PDSI

SPI03

SPI24 SPI09

26 JJA

cor with NINO3.4 SST

(27)

ENSO-drought connection: Roles of soil moisture

27 cor with DJF NINO3.4 SST

Soil moisture Evaporation

(28)

ENSO-drought

connection: Roles of soil moisture

28

Soil Moisture Evaporation NINO3.4

SST DJF

DJF MAM JJA DJF MAM JJA P GP

DJF

0.23 0.40 0.46 0.34 0.16 0.47 0.31 P SW

DJF

0.36 0.61 0.71 0.55 0.20 0.67 0.60

ENSODJF P

DJF-MAM Soil

MAM-JJA Evap

JJA drought

Great Plains

Southwest

(29)

SST-drought relationship

Subtropical North Atlantic (SNA)

PDSI

SPI03

SPI09

SPI24

29 JJA

cor with SNA SST

DJF SNA SST JJA SNA SST

(30)

SST-drought relationship

PDSI

SPI03

SPI09

SPI24

30 JJA

cor with TIO SST

DJF TIO SST JJA TIO SST

(31)

Types of SST-drought relationship

P simultaneous (JJA) SST SST leading to summer P

P --- persistent --- P persisting SST

SST --- persisting --- SST leading to persistent P P SM P preceding (DJF) SST

SST leading to summer P via SM

P --- persistent --- P persisting SST leading to persistent P

SST --- persisting --- SST preceding SST leading to JJA P via SM

SM

DJF JJA

31 short-term drought

persistent drought I

II

I+II

(32)

Auto-persistence of SPI and roles of soil moisture and evaporation

SM(DJF-MAM)

P(DJF)

32 How to distinguish them?

SPI(DJF)

(1) Auto-persistence larger than cor (SM-JJA SPI)

Possibilities:

SM(DJF-MAM)

SPI(JJA) SPI(JJA)

P(DJF) SPI(DJF)

(2) Cor (SM-JJA SPI) larger than auto-persistence

(3): (1) + (2)

(33)

Findell & Eltahir’97

QuickTime™ and a decompressor

are needed to see this picture.

Precipitation persistence vs soil moisture on precipitation

33

(34)

Table 1. Auto-correlations of drought indices from winter to summer and correlations between the winter or spring soil moisture and the summer drought indices for the Great Plains during 1932-2005.

Auto-correlation Soil (DJF)-drought (JJA) Soil (MAM)-drought (JJA)

PDSI 0.71 0.67 0.85

SPI01 0.27 0.27 0.37

SPI02 0.28 0.26 0.42

SPI03 0.26 0.21 0.44

SPI06 0.26 0.25 0.61

SPI09 0.31 0.44 0.74

SPI12 0.51 0.71 0.86

SPI24 0.83 0.85 0.84

The larger correlation between MAM soil moisture and JJA drought than the auto-correlation of drought from DJF to JJA indicates the contribution of soil moisture to the persistence of droughts from DJF to JJA.

SPI persistence vs soil moisture on drought

34

(35)

Rajagopalan et al. 2000

ENSO-drought relationship:

Long-term Change

Cor (Summer PDSI-Winter NINO3)

35

1895-1928 1929-1962

1963-1995

(36)

Cole and Cook’98

QuickTime™ and a decompressor

are needed to see this picture.

Cor (DJF SOI-Annual PDSI)

36 ENSO-drought relationship:

Long-term Change

(37)

DJF NINO3.4 SST

37 strong

weak

strong SW SPI

GP SPI Larger fluctuations in GP than in SW

(38)

JJA NINO3.4 SST

38 SW SPI

GP SPI

weak/negative

strong/positive

(39)

DJF TIO SST

39 SW SPI

GP SPI

(40)

JJA TIO SST

40 weak

strong

SW SPI

GP SPI weak

strong

(41)

Summary

The relationship of boreal summer U.S. droughts with SST and soil moisture differs significantly between short-term and long-term

droughts. The short-term droughts (<= 3 months) are mostly

influenced by simultaneous SST forcing. The medium-term and

long-term droughts (>= 6 months) are influenced by both preceding and simultaneous SST forcing. The soil moisture change shows

obvious leading for medium-term and long-term droughts.

A dominant remote forcing for U.S. droughts is tropical Pacific SST.

Tropical Indian Ocean SST forcing has notable influence on

medium-term and long-term droughts. Additional impacts for short- term and medium-term droughts are from the North Atlantic SST forcing.

41

(42)

Summary (continued)

The most notable impacts of the tropical Pacific SST forcing on medium-term and long-term droughts are found in the Southwest

with extension to the Great Plains. Anomalous soil moisture induced by remote ENSO forcing contributes to the persistence of droughts from winter to summer through anomalous evaporation during late spring to summer.

The relationship between tropical Pacific SST and boreal summer U.S. droughts show obvious long-term changes. In comparison, the long-term change is more pronounced for the GP droughts than for the SW droughts. Obvious long-term changes are also found in the correlation of U.S. droughts with tropical Indian Ocean SST,

especially for JJA SST.

42

(43)

THANK YOU!

43

Referenties

GERELATEERDE DOCUMENTEN

Figuur 1 geeft het verloop van de pootscore weer voor de periode juli 1999 tot december 2001.. Het verloop van de pootscore laat over de afgelopen jaren een

Keywords: Mixed lubrication; deep drawing; asperity flattening; friction modeling;.. *

That the world faces an epidemic of chronic diseases is unmistakable. Chronic diseases represent 63% of all deaths worldwide. Against the background of demo- graphic aging,

These deposits are abnormally hyperphosphorylated tau protein, known as neurofibrillary tangles (NFT), which cause intracellular changes. These NFTs have a specific

het, sal dit ook die on d ergang wees van die Afrlkanervolk en Suid-Afrika, indicn nie betyds die Afrikaner daartoc kom te ontwaak nic, om te besef watter

Waarnemings: Gebruik hierdie simbole om ekstra inligting by te voeg.. Observations: Use these symbols to add additional information to

regeling der Commissie te maken, ma~r alleen wordt op voorstel van Ouderling Hatting, ~esecondeerd door Diaken Engelbrecht, vastge.steld, dat de Commissie met haren

In Figure 31 we find the probability distribution of the characteristic years based on the different indices, combined with the expected damage in those years.. The results would