AM”I, Tel-Aviv, 22 March 2011
description
Transcript of AM”I, Tel-Aviv, 22 March 2011
AM”I, Tel-Aviv, 22 March 2011
סימולציות אקלימיות של שינויים במאזן המים מעל אגן הים התיכון עם מבט
לישראל – האם אנו מתייבשים?
פינחס אלפרט
ביה"ס פורטר לסביבה והחוג לגיאופיסיקה,
אוניברסיטת תל אביב
בשיתוף עם:ג'ין פנגג'ון, שמעון קריצ'ק, רנה סמואלס, יצחק כרמונה,מעין הראל
Overview
• Factor Separation and synergy- New Book
• Super High-Resolution MRI 20-km climate model- new results
• High Resolution Climate Modeling over the Mediterranean
• High Resolution Model Ensemble over the E. Mediterranean
• Conclusions
PublishedFeb. 2011Cambridge Univ.
Press
One comment on the importance of a proper Factor Separation;
from the Book’s Introduction
“There seems to be some basic psychological tendency in human thinking to present results linearly, with the hidden assumption that the synergies are small or can be ignored.
Nonlinearities in the atmosphere, however, are often significant, and therefore need to be calculated and separated from the pure contributions of each factor, as shown in many very different applications presented in this Book.”
High Resolution Model Ensemble over the E. Mediterranean
• 4 climate models (18-25 km horizontal resolution)
– Japanese MRI 20km– ECHAM- RegCM- 25 km– ECHAM- MM5- 18 km– HADLEY-MM5- 18 km
– First, some results from the Super High-Resolution MRI 20-km climate model:
A. Kitoh, A. Yatagai and P. Alpert, "First super-high-resolution model projection that the ancient “Fertile Crescent” will disappear in this century", Hydrolo. Res. Lett., 2, 1-4, DOI: 10.3178/HRL.2.1, 2008.
Annual Precipitation (mm/year)
CRU
EMclimIPCC AR4 models
20-km model
Total seasonal (Oct-Apr) average precipitation for 1979-2002
The six stations are from south-to-north, Egypt---Cairo (Ca,); Israel---Beer-Sheva (Bs), Tel-Aviv (Ta), Haifa (Hf); Lebanon---Beirut (Be) and Turkey---Adana (Ad). Unit: mm/season.
Comparison of average total observed seasonal P
P E
West inflowWest outflow
S-InF S-OutF
E-InF
E-OutF
N-InF N-OutF
Sketch map
P Categories12345
Amount P P<11=<P<1.51.5=<P<22=<P<2.5 P<=2.5
Unit: mm/day
Water vapor budget equation
A
IF
A
OFdpdlqvdp
g
T
S n
T
S
Q1
A
IF
A
OFPE
Using Green’s Theorem
PEdpg
dpg
qdptg C
T
S
T
S
T
S
QQ111
dPW Vertical integrated moisture flux (VIMF)
VICLFCL=cloud Liquid
By neglect dPW, VICLF +
Area
Total boundary outflow and
inflow
Difference of seasonal total E, P and P-E between the future (2075-2099) and current (1979-2002) 20 km GCM runs. Dashed contour lines indicate the negative changes, i.e. reduction in the future. Unit: mm/season
Annual anomaly of P(black bar) and E (shaded bar) for the future (2075-2099), separated by sea and land area over the Mediterranean region. Calculated by individual year of the future minus the current mean (1979-2007). Unit: mm/day.
Annual P,E changes during 2075-2099 compared to current mean
E
P
Precipitation
categories
CurrentFuture
monthspercentageAver_PmonthspercentageAver_P
P<175%0.901913%0.75
1.5<p>=14124%1.293323%1.26
2<p>=1.55432%1.766142%1.72
2.5<p>=24627%2.242316%2.17
P>=2.52012%2.7286%2.75
Sum168
)28 y(100%1.85
)Mean P(
144)24 y(
100%1.62)Mean P(
Five precipitation categories based on monthly averages in (mm/d) over the whole Mediterranean- current & future
Relationships among the moisture budget components based on the 5 different precipitation categories
ΔP
Unit: mm/day
Current (1979-2007)
Future (2075-2099) minus current
Tinf Tout P E E-P Tout-Tinf
Inflow Outflow
W E N SW
E
N
S
W E N SW E N
S
Comparison of monthly mean precipitation from observed rain gauge (black column) and 20km GCM (grey column) for selected six stations in Israel based on their locations. Unit: mm/day. The selected stations are: Northern Israel---(a) Har-Knaan, (b) Eilon; Center Israel---(c) Tel-Aviv, (d) Jerusalem; Southern Israel---(e) Beer-Sheva, (f) Elat.
Har-Knaan
Tel-Aviv
Beer-Sheva
Streamflow 20km present MRI SST
MIROC SST
in (m3/s)2081-2100 changes
Changes of monthly mean river discharge of six rivers by (1979-2003) compare to (2075-2099). Except to the Jordan River, all rivers flow into the Mediterranean (m3/s). Bold lines ( ) are for current climate, while dashed ( ) for the future.
Early RCM Transient Run A1B 1960-2060
50 km
P. Alpert, S.O. Krichak, H. Shafir , D. Haim, and I. Osetinsky, "Climatic trends to extremes employing regional modeling and statistical interpretation over the E. Mediterranean", Global and Planetary Change, 63, 163-170, 2008.
Annual Precipitation (mm)
0
200
400
600
800
1000
1200
1400
1600
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
2005
2008
2011
2014
2017
2020
2023
2026
2029
2032
2035
2038
2041
2044
2047
2050
2053
2056
2059
mm
RCMEin Ziw anMaayan Baruch
-~30%
7-day dry spells
0
2
4
6
8
10
12
14
16
18
20
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
2005
2008
2011
2014
2017
2020
2023
2026
2029
2032
2035
2038
2041
2044
2047
2050
2053
2056
2059
tim
es
RCMEin Ziw anMaayan Baruch
3-day wet spells
0
2
4
6
8
10
12
14
16
18
20
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
2005
2008
2011
2014
2017
2020
2023
2026
2029
2032
2035
2038
2041
2044
2047
2050
2053
2056
2059
tim
es
RCMEin Ziw anMaayan Baruch
First High Resolution Climate Model Ensemble over the E. Mediterranean
• 4 climate models (18-25 km horizontal resolution)
– Japanese MRI 20km– ECHAM- RegCM- 25 km– ECHAM- MM5- 18 km– HADLEY-MM5- 18 km
• SRES A1B scenario• Weighted ensemble based on Jensen-Shannon
Divergence metric– The models with higher similarity to observations over
a control period are given higher weights
Example: Kfar Giladi-Example: Kfar Giladi- Total Annual Rainfall in mm/y – Total Annual Rainfall in mm/y –
Two past periods (1965-94; 1970-99)Two past periods (1965-94; 1970-99)
Probability Density Function for Kfar Giladi for observed data (2 time periods :
1965-1994) solid ,(1970-1999 (dashed), and 4 models .
p
mm/y
OBS
Example: Kfar Giladi-Example: Kfar Giladi- Number of Wet Days per season Number of Wet Days per season
Two past periods (1965-94; 1970-99) Two past periods (1965-94; 1970-99)
Probability Density Function for Kfar Giladi for observed data (2 time periods :
1965-1994) solid ,(1970-1999 (dashed)), and 4 models. Wet Days shown.
p
Number of Days
Example: Kfar Giladi-Example: Kfar Giladi- Number of 3 Day Wet Spells Number of 3 Day Wet Spells
Two past periods (1965-94; 1970-99) Two past periods (1965-94; 1970-99)
Probability Density Function for Kfar Giladi for observed data (2 time periods :
1965-1994) solid ,(1970-1999 (dashed)), and 4 models. Wet Spells shown.
p
Number of spells
0 500 1000 1500 20000
1
2x 10
-3 1965-1994
0 500 1000 1500 20000
1
2x 10
-3K
efar
Gila
di
0 500 1000 15000
2
4x 10
-3
0 500 1000 15000
2
4x 10
-3
Tel
Avi
v
0 500 1000 15000
2
4x 10
-3
0 500 1000 15000
2
4x 10
-3
Jeru
sale
m
0 100 200 300 400 5000
0.005
0.01
0 100 200 300 400 5000
0.005
0.01
Bee
r S
hev
a
0 500 1000 1500 20000
1
2x 10
-3 2035-2060
0 500 1000 15000
2
4x 10
-3
2015-2035
0 500 1000 15000
2
4x 10
-3
0 100 200 300 400 5000
0.005
0.01
Change in JSD calculated PDF over time for Change in JSD calculated PDF over time for Average Annual AmountsAverage Annual Amounts
p
mm/yr
shows observed data (red) and calculated past
near future (2015-2035 - dashed) and past calculated
(1965-1990 - solid).
far future (2035-2060 - dashed) and past calculated
(1965-1990 - solid).
0 50 100 1500
0.01
0.02
0.03
0.04
Kfa
r G
ilad
i
1965-1994
0 50 100 1500
0.01
0.02
0.03
0.042015-1035
0 50 1000
0.01
0.02
0.03
0.04
Tel
Avi
v
0 50 1000
0.01
0.02
0.03
0.04
0 50 1000
0.01
0.02
0.03
0.04
Jeru
sale
m
0 50 1000
0.01
0.02
0.03
0.04
0 20 40 60 800
0.02
0.04
0.06
Bee
r S
hev
a
0 20 40 60 800
0.02
0.04
0.06
0 50 100 1500
0.01
0.02
0.03
0.042035-2060
0 50 1000
0.01
0.02
0.03
0.04
0 50 1000
0.01
0.02
0.03
0.04
0 20 40 60 800
0.02
0.04
0.06
Change in JSD calculated PDF over time for Change in JSD calculated PDF over time for Number of Wet DaysNumber of Wet Days
p
Number of Days
shows observed data (red) and calculated past
near future (2015-2035 - dashed) and past calculated
(1965-1990 - solid).
far future (2035-2060 - dashed) and past calculated
(1965-1990 - solid).
0 5 10 15 20 250
0.05
0.1
0.15
0.2
Kfa
r G
ilad
i
1965-1994
0 5 10 15 20 250
0.05
0.1
0.15
0.22015-2035
0 5 10 15 200
0.05
0.1
0.15
0.2
Tel
Avi
v
0 5 10 15 200
0.05
0.1
0.15
0.2
0 5 10 150
0.1
0.2
0.3
Jeru
sale
m
0 5 10 150
0.1
0.2
0.3
0 5 10 15 20 250
0.05
0.1
0.15
0.22035-2060
0 5 10 15 200
0.05
0.1
0.15
0.2
0 5 10 150
0.1
0.2
0.3
Change in JSD calculated PDF over time for Change in JSD calculated PDF over time for Number of Wet SpellsNumber of Wet Spells
p
Number of Spells
shows observed data (red) and calculated past
near future (2015-2035 - dashed) and past calculated
(1965-1990 - solid).
far future (2035-2060 - dashed) and past calculated
(1965-1990 - solid).
2002-1976
1975-1948
Heat waves increase & longer
Frequency of hot & cold spells in July-Aug in both halfs of 1948-
2002
B. Ziv , H. Saaroni, A. Baharad, D. Yekutieli and P. Alpert, "Indications for aggravation in Summer heat conditions over the Mediterranean basin", Geoph. Res. Lett., 32, L12706, doi:10.1029/2005GL022796, 2005.
Durations (days) of hot/cool spells
Distribution of 850 hPa daily temperatures
for 1948-1977 and 1973-2002
Shift in the mode temperatures Increase in extreme events
Changes in T-850 mb distributions over the E. Mediterranean
• Most frequent value increases• Distribution widens• Increase in heat waves & their intensity
B. Ziv , H. Saaroni, A. Baharad, D. Yekutieli and P. Alpert, "Indications for aggravation in Summer heat conditions over the Mediterranean basin", Geoph. Res. Lett., 32, L12706, doi:10.1029/2005GL022796, 2005.
Climatic TrendsClimatic Trends
"Trends in Middle East climate extreme indices from 1950 to
2003",
X. Zhang, E. Aguilar, S. Sensoy, H. Melkonyan, U. Tagiyeva, N. Ahmed, N. Kutaladze, F. Rahimzadeh, A. Taghipour, T. H. Hantosh, P. Alpert, M. Semawi, M. K. Ali, M. H. S. Al-Shabibi, Z. Al-Oulan, T. Zatari, I. Al Dean Khelet, S. Hamoud, R. Sagir, M. Demircan, M. Eken, M. Adiguzel, L. Alexander, T. C. Peterson, and T. Wallis,
"Trends in Middle East climate extreme indices from 1950 to 2003", J. Geophys. Res., 110, D22104, doi:10.1029/2005JD006181, 2005
TX90p
TN90p
Top 10PercentilesTmax/min
Recent20 y dramatic
TX90p
TN90p
Top 10PercentilesTmax/min
Solid triangles5% significanceTurkey- recent & TN More significant
Annual Precipitation Anomalies
Precipitation variation is characterized by strong interannual variability without any significant trend in any of the indices
Mediterranean Changes in P: 1980-2002 vs. 1931-79
Actual change in annual means
(mm)Data from UEA TS2p1
t-test (95% confidence highlighted)
Y. Kushnir 2009
Max 1 day Precipitation Amount
Annual TotalPrecipitation
Zhang et alJGR, 2005
Middle East- Mixed trendsUnlike most Mediterranean
Conclusions
• Tendency to extreme in daily and seasonal rainfall- Desertification?
• Tendency to extreme temperatures• Japanese run confirms how crucial are
high-resolution climate runs• First High Resolution Model Ensemble
over the E. Mediterranean confirms the tendency to extreme along with less total rainfall
Conclusions• The 20km GCM data better represents the precipitation
region over the Middle east region, an increasing evaporation with the magnitude of 150-200 mm/season at the water bodies of EM are projected; a significant decrease of precipitation was found over west Turkey, west Syria, entire Israel and Lebanon, with a magnitude of over 200 mm/wet seasonat the end of this century. The famous “Fertile Crescent” precipitation strip becomes much drier.
• For the current climate, the E of the eastern Mediterranean sea (EMS) is higher than that of western Mediterranean sea (WMS) with a average value of 0.4 mm/day, but with the precipitation it is opposite, i.e. less than in the WMS with an average value of 0.32 mm/day. For the future, the evaporation increases for the EMS are higher than that of WMS, with the average value of 0.45 and 0.22 mm/day respectively; while the precipitation decreases for the WMS are higher than that of theEMS, with the average value of -0.21 and -0.16 mm/day.
• The separated calculated boundary moisture flux balance the moisture budget equation quite well, the main source of moisture for the precipitation over the Mediterranean comes from the west and north boundaries. The evaporation acts a significant positive role to the precipitation of Mediterranean.
• Even the numbers of precipitation events decrease in the future, however, the absolute value for the largest precipitation category shows a increasing trend.
• The decrease of river flow, increase evaporation and decrease precipitation make the water crisis available over the study area at the end of this century.