WEG Automatic Voltage Regulator Grt7 Th4 r2 10001284109 Manual English
TH4.TO4.5.ppt
-
Upload
grssieee -
Category
Technology
-
view
296 -
download
0
description
Transcript of TH4.TO4.5.ppt
2011 IEEE Geoscience and Remote Sensing symposiumJul. 24-29, Vancouver, Canada
Analysis of the passive microwave high-frequency signal in the shallow snow retrieval
[email protected] 2828th th Jul. 2011Jul. 2011
AMSR-E
Y.B. Qiu1*, H.D. Guo1, J.C. Shi2, S.C. Kang3
J. Lemmetyinen4, J.R. Wang5
1 Center of Earth Observation and Digital Earth, CAS, China2 University of California, Santa Barbara, USA3 Institute of Tibetan Plateau Research, CAS, China4 FMI, Arctic Research Centre, Finland5 NASA Goddard Space Flight Center, Greenbelt, USA
OutlineSnow Products Analysis in China
Why Snow - a potentially sensitive factor of climate change – debate ? Need more accuracy snow products over China area
Passive Microwave remote sensing of snow Nowadays, the operational algorithm regime– Gradient
(36/18GHz) Shallow snow situation in China
Possibility analysis of the high frequencies in the shallow snow retrieval
Comparison: In-situ snow depth and SMM/I, and AMSR-E emission signal
Possible algorithm development with high frequencies and analysis
Conclusion
Snow - a potentially sensitive factor of climate change
Snow – very important IPCC AR4(2007): Continental-scale snow cover extent (SCE)
is a potentially sensitive indicator of climate change. (Foster et al. 1982; Namias 1985; Gleick, 1987) said: Snow
is not only a sensitive indicator of climate change, but makes feedbacks to it.
Snow has been proposed as a useful indicator in testing and monitoring global climate change (Robinson et al. 1990).
…
Works support IPCC-AR4 ?Nine GCM model: shrinking snow cover over Northern
American
( Frei, A. and G. Gong, 2005. )
Why Snow?
Will Climate Change Affect Snow Cover Over North America?
North American snow models miss the mark – observed trend opposite of the predictions
Goddard uses data from Rutger’s Global Snow Lab to claim that the latest 22-year trend for Winter (Dec, Jan, Feb) in the Northern Hemisphere invalidates the CMIP3 modeling of snow extent as presented by Frei and Gong in 2005.
Why Snow? debate
New debate ?
Snow Cover Distribution, Variability, and Response to Climate Change in Western China
Data :SMMR-SD, NOAA-SCA Results show that western China did not experience a
continual decrease in snow cover during the great warming period of the 1980s and 1990s. The positive trend of the western China snow cover is consistent with increasing snowfall, but is in contradiction to regional warming.
Potential impact of climate change on snow cover area in the Tarim River basin
QIN DAHE, 2006
Xu Changchun, 2007
Data: 1982–2001, station data The SCA of the entire basin showed a slowly increasing
trend. Correlation analysis implied that the SCA change in the
cold season was positively correlated with the contemporary precipitation change, but had no strong correlation with the contemporary temperature change.
China – Publication’s ViewChina – Publication’s View
Western China
Tibet Plateau
Northern China
XinjiangMongolia
Northeast
The select test areas in China
Data: Rutger snow productSSM/I SWE productNOAA IMS 4Km/24Km
Part I: Analysis in China from the nowadays products:
Qinghai
Xinjiang
TibetQinghai
Tibet
1966-11-1 1969-11-1 1972-11-1 1975-11-1 1978-11-1 1981-11-1 1984-11-1 1987-11-1 1990-11-1 1993-11-1 1996-11-1 1999-11-1 2002-11-1 2005-11-1 2008-11-10
20000400006000080000
100000120000140000160000180000200000220000240000260000280000300000320000340000360000380000400000
Sq
ua
re k
m
Date
1966-11-1 1969-11-1 1972-11-1 1975-11-1 1978-11-1 1981-11-1 1984-11-1 1987-11-1 1990-11-1 1993-11-1 1996-11-1 1999-11-1 2002-11-1 2005-11-1 2008-11-10
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
55000
60000
Sq
ua
re k
m
Date
Equation y = a + b
Adj. R-Squar 0.02373
Value Standard Err
Book2_E Intercept -325738.69 89219.84807
Book2_E Slope 0.13331 0.03645
1966-11-11969-11-1
1972-11-11975-11-1
1978-11-11981-11-1
1984-11-11987-11-1
1990-11-11993-11-1
1996-11-11999-11-1
2002-11-12005-11-1
2008-11-1
0
400000
800000
1200000
1600000
2000000
2400000
2800000
Sq
ua
re k
m
Date
Western China Area
Tibet Plateau
Northern China
Agree well with the previous publication, but the snow cover area over Tibet is not quite right.
Rutger snow product
1966.11~2010.5
7811017911018011018111018211018311018411018511018611018711018811018911019011019111019211019311019411019511019611019711019811019911010011010111010211010311010411010511010611010711010
100000
200000
300000
400000
500000
SW
E(m
m)
Date
China – decreasing SCA/SWE over Tibet Plateau
SMM/I SWE product
7811017911018011018111018211018311018411018511018611018711018811018911019011019111019211019311019411019511019611019711019811019911010011010111010211010311010411010511010611010711010
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
45000000
SW
E(m
m)
Date
Tibet Plateau
Tibet Plateau
1978-2007
NOAA IMS
1997-2010 SCA
970201970801
980201980801
990201990801
000201000801
010201010801
020201020801
030201030801
040201040801
050201050801
060201060801
070201070801
080201080801
090201090801
100201
0
1000000
2000000
3000000
4000000
5000000
6000000
Squ
are
km
Date
Tibet Plateau
Western China
decreasing
decreasing
970201 970801 980201 980801 990201 990801 000201 000801 010201 010801 020201 020801 030201 030801 040201 040801 050201 050801 060201 060801 070201 070801 080201 080801 090201 090801 1002010
1000000
2000000
3000000
4000000
Sq
ua
re k
m
Date
Equation y = a +
Adj. R-Sq 0.0081
Value Standard E
B Interce 1.14445 1.79336E7
B Slope -46.201 7.31125
970201 980201 990201 000201 010201 020201 030201 040201 050201 060201 070201 080201 090201 1002010
1000000
2000000
3000000
4000000
5000000
6000000
Squ
are
km
Date
D Linear Fit of D
Equation y = a + b*x
Adj. R-Squar 3.55109E-
Value Standard Error
D Intercept -7.46397E 4.68434E7
D Slope 31.09071 19.09734
NOAA IMS
1997-2010 SCA Northern China
The NOAA IMS show different Trend with above two publications and the Rutger’s result.
Increasing
We get:Different products provide different time-series appearance on the snow factorThe SCA from Rugter is not quite right over Tibet China.SWE is a quite valuable parameter for its long times series records (SMMR, SMM/I)
Need inter-comparison and validation of certain snow cover products.
We get:
In all,According to the analysis, snow cover over Tibetan Plateau is quite different with other places over Northern hemisphere, We need accuracy estimation of the snow cover parameters for a long time to convince the climatology analysis to corresponding the global environment change research.
Need More accurate snow data
Part II: Passive Microwave remote sensing of snow
Snow emission model – understanding the snow microwave emission
DMRT – theoretical… MELMES – multi-layer model HUT Snow Emission Model
– 2010 now extent to multilayer modal …
Snow Temperature ProfileSnow grain size profileSnow density profileSnow depth (SD)Interface roughness
For dry snow
Passive Microwave remote sensing of snow
Algorithms Basically, base on the satellite brightness temperature (TB) difference
between 36GHz and 18GHz Goodison & Walker,1995:SMMR & SSM/I, TB(19V-37V) Goita et al.,1997: forest area TB(19V-37V) Kelly,Chang,Foster,& Hall,2001: second order of TB(19V-37V) Pulliainen , & Hallikainen,2001, iteration algorithm (match) …
0 20 40 60 80 100 120 140 160 180 200 220 240-140
-120
-100
-80
-60
-40
-20
0
TB
Diff
ere
nce
(K)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.9036.5-18.7
TB(37-19V) (K)
Station Snow Depth (cm)
Shallow snow situation in China – western China, especially the Tibet area
1 3 5 7 9 11 13 15 17 19 21 23 250
1
2
3
4
5
Sno
w D
epth
(cm
) Snow Depth (cm)
DanXung Station 1980-1 Snow Evolution
Jan 7 Jan 12 Jan 17 Jan 22 Jan 27 Feb 1 Feb 60
2
4
6
8
10
12
Sno
w D
epth
(cm
)
Snow Depth (cm)
DanXung Station 1981 Snow Evolution
Menyuan station Maduo station
shallow snow over Tibet area<15cm or 20cm
Sparse Station
Shallow snow situation in China – western China
2009-9-1 2009-11-1 2010-1-1 2010-3-1 2010-5-1
-40
-30
-20
-10
0
10
0
3
6
9
12
15
Sn
ow
De
pth
(cm
)
89V-18V 36V-18V 18V-10.7V
TB
Diff
ere
nce
(K)
Date
Snow Depth(cm)
52985 Hezheng, Gansu (35.42N, 103.33E)
2009-9-1 2009-11-1 2010-1-1 2010-3-1 2010-5-1
-40
-30
-20
-10
0
10
0
3
6
9
12
15
Sn
ow
De
pth
(cm
)
89V-18V 36V-18V 18V-10.7V
TB
Diff
ere
nce
(K)
Date
Snow Depth(cm)
Snow depth < 10cm
56093 Minxian, Gansu (34.43N, 104.02E)
Part III: Possibility analysis of the high frequencies in the shallow snow retrieval
HUT snow model simulation
0 20 40 60 80 100 120 140 160 180 200 220 240-140
-120
-100
-80
-60
-40
-20
0
TB
Diff
ere
nce
(K)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.9036.5-18.7
0 20 40 60 80 100 120 140 160 180 200 220 240-200
-180
-160
-140
-120
-100
-80
-60
-40
-20
089.0-18.7
TB
Diff
eren
ce(K
)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.90
36GHZ/18.7GHz gradientShallow snow sensitive
89GHZ-18.7GHz gradientShallow snow sensitive
Comparison: In-situ snow depth and SMMR, SMM/I, and AMSR-E emission signal
Snow depth (cm) the Former Soviet 284 station records V2.0 (1966~1996) The snow depth (cm) over China (2009.9~2010.5) NamCo station snow measurement for one whole winter
(2007~2008)
Comparison of the traditional algorithm records and Snow depth
Satellite datasetSMMR(1978~1987), SSM/I (1987~1996), AMSR-E swath data
Tree-free, sparse vegetation area, taken at the winter of 2010
Less than 20cm
NamCo station
Snow depth (cm) from NamCo station and AMSR-E TBs
Compare with the NASA algorithm SWE
result
TB(35GHz-18.7GHz)
TB(89GHz-18.7GHz)
Snow depth (cm)
SWE (mm)/NASA
ITPR,CAS
Snow depth (cm) and AMSR-E TBs (2009~2010)
50727 (47.17N 119.93E) Heilongjiang
50434 50.48N 121.68ENeimenggu
Snow depth (cm) and AMSR-E TBs (2009~2010)
54049 44.25N 123.97E Jilin
52681 38.63N 103.08E Gansu
Snow depth (cm) and AMSR-E TBs (2009~2010)
53231 41.40N 106.40ENeimenggu
53519 39.22N 106.77ENingxia
89GHz is sensitive to the snow
occurrence (fresh snow flake)
Snow depth (cm) and SSM/I TB Gradient
The select station, the altitude > 1500mNear Tibet, ChinaAlmost the shallow snow cover areaSparse forest
Snow depth and SSM/I TB Gradient
36665 ZAJSAN
0
10
20
30
40
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-80
-70
-60
-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
36870 ALMA-ATA
0
10
20
30
40
50
60
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-70
-60
-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
TB(37-19GHz) and TB(85-19GHz) ‘s response to the snow evolution
38618 FERGANA
0
10
20
30
40
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
38599 LENINABAD
0
10
20
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-70
-60
-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
High frequency shows a sensitive response to the shallow snow
Snow depth and SSM/I TB Gradient
SAMARKAND 38696
Snow depth and SSM/I TB
0
10
20
30
40
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-70
-60
-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
KURGAN-TJUBE 38933
0
10
20
30
40
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996-70
-60
-50
-40
-30
-20
-10
0
10
20
Tb
Diff
ere
nce
(K)
Snow Depth
Sn
ow
De
pth
(cm
)
37V-19V
Year
85V-19V 85V-37V
High frequency shows a sensitive response to the shallow snow
Possible algorithm development with high frequencies and analysis
We apply the ATC Chang(1987) gradient algorithm
SD = a* (Tb18H - Tb37H) + bFirstly, we exam the relatively deep snow over rich forest area.
1987-1991 snow depth and SMM/I Tbs, 261 samples
0 5 10 15 20 25 30 35 40 45 50 55-30
-25
-20
-15
-10
-5
0
TB
diff
ere
nce
(K)
Snow Depth(cm)
37-19V Linear Fit of 37-19V
R-square: 0.71428a=-6.98161 b= -
0.35971
Possible algorithm development with high frequencies and analysis
1987-1991 snow depth and SMM/I Tbs, 261 samples
R-square: 0.05764a=-18.95129 b= -
0.12906
0 5 10 15 20 25 30 35 40 45 50 55
-40
-30
-20
-10
0
TB
diff
eren
ce(K
)
Snow Depth(cm)
85-19V Linear Fit of 85-19V
0 20 40 60 80 100 120 140 160 180 200 220 240-200
-180
-160
-140
-120
-100
-80
-60
-40
-20
089.0-18.7
TB
Diff
ere
nce
(K)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.90
0 5 10 15 20 25 30 35 40 45 50 55-30
-25
-20
-15
-10
-5
0
5
10
15
TB
diff
ere
nce
(K)
Snow Depth(cm)
85-37V Linear Fit of 85-37V
R-square: 0.2509a= -11.96944 b=
0.23074
Grain Size =1.3~1.7mm
Possible algorithm development with high frequencies and analysis
1992-1995 snow depth and SMM/I Tbs, 640 samples
R-square: 0.48892a= -6.24384 b= -
0.26904
0 5 10 15 20 25 30 35 40 45 50 55 60 65-30
-25
-20
-15
-10
-5
0
TB
diff
ere
nce
(K)
Snow Depth(cm)
TB difference(K) Linear Fit of 37-19V
Possible algorithm development with high frequencies and analysis
1992-1995 snow depth and SMM/I Tbs, 640 samples
R-square: 0.00456a= -17.45683 b= -
0.03825
R-square: 0.33649a= -11.21299 b=
0.23079
0 5 10 15 20 25 30 35 40 45 50 55 60 65-40
-20
0
TB
diff
eren
ce(K
)
Snow Depth(cm)
89-19V Fit of 89-19V
0 5 10 15 20 25 30 35 40 45 50 55 60 65-20
-15
-10
-5
0
5
10
15
TB
diff
ere
nce
(K)
Snow Depth(cm)
85-37V Fit of 85-37V
Thick snow
Thick snow
Possible algorithm development with high frequencies and analysis
1988-1995 snow depth and SMM/I Tbs, 420 samples
0 5 10 15 20 25-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
TB
diff
ere
nce
(K)
(37
-19
V)
Snow Depth(cm)
TB difference(K) (37-19V) Linear Fit of TB difference(K)
R-square:0.31795a= -6.66873 b=-
1.07606
Shallow snow
Possible algorithm development with high frequencies and analysis
0 5 10 15 20 25-70
-60
-50
-40
-30
-20
-10
0
10
TB
diff
ere
nce
(K)
(85
-19
V)
Snow Depth(cm)
TB difference(K) (85-19V) Linear Fit of TB difference(K)
R-square:0.36473a=-19.22 b=-
1.5358
Select area –Tibet Plateau
Possible algorithm development with high frequencies and analysis
1987-1995 snow depth and SMM/I Tbs, 469 samples
0 5 10 15 20 25 30 35 40 45-50
-40
-30
-20
-10
0
10
TB
diff
eren
ce(K
) (3
7-19
V)
Snow Depth(cm)
TB difference(K) (37-19V) Linear Fit of TB difference(K)
0 5 10 15 20 25 30 35 40 45-80
-60
-40
-20
0
20
TB
diff
eren
ce(K
) (8
5-19
V)
Snow Depth(cm)
TB difference(K) (85-19V) Linear Fit of TB difference(K)
0 5 10 15 20 25 30 35 40 45-60
-30
0
TB
diff
ere
nce
(K)
(85
-37
V)
Snow Depth(cm)
TB difference(K) (85-37V) Linear Fit of TB difference(K)
0 20 40 60 80 100 120 140 160 180 200 220 240-200
-180
-160
-140
-120
-100
-80
-60
-40
-20
089.0-18.7
TB
Diff
ere
nce
(K)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.90
Possible algorithm development with high frequencies and analysis
1987-1995 snow depth and SMM/I Tbs, 1353 samples
0 10 20 30 40 50 60 70-60
-30
0
TB
diff
ere
nce
(K)
(37
-19
V)
Snow Depth(cm)
TB difference(K) (37-19V) Linear Fit of TB difference(K)
0 30 60
-80
-40
0
TB
diff
ere
nce
(K)
(85
-19
V)
Snow Depth(cm)
TB difference(K) (85-19V) Linear Fit of TB difference(K)
The pair 85.0/19 shows its shallow snow retrieval ability and when the snow depth over 20cm, the signal is more variable and suspect.
Possible algorithm development with high frequencies and analysis
Atmosphere influence issue
HUT snow emission model (Satellite and Ground)
0 20 40 60 80 100 120 140 160 180 200 220 240
-20
0
TB
Diff
eren
ce(K
)
SWE(mm)
Snow Grain Size(mm)
0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.9036.5-18.7
0 20 40 60 80 100 120 140 160 180 200 220 240-140
-120
-100
-80
-60
-40
-20
0
TB
Diff
ere
nce
(K)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.9036.5-18.7
Ground
Satellite
Tb(36.5GHz-18.7GHz)
Atmosphere influence analysis
HUT snow emission model (Satellite and Ground)
0 20 40 60 80 100 120 140 160 180 200 220 240
-100
-80
-60
-40
-20
0
TB
Diff
eren
ce(K
)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.9089.0-18.7
0 20 40 60 80 100 120 140 160 180 200 220 240-200
-180
-160
-140
-120
-100
-80
-60
-40
-20
089.0-18.7
TB
Diff
eren
ce(K
)
SWE(mm)
Snow Grain Size 0.5 0.7 0.9 1.10 1.30 1.50 1.70 1.90
89.0GHz-18.7GHzSatellite
Ground
Atmosphere influence
Conclusion
Snow parameter is a critical climate indicator.Over western China, the nowdays snow products provide different trend regionally.The high frequencies is sensitive to the snow appearance and could be good at the beginning of the snow accumulation.Over relative deep snow (> 20cm) the Tbs at 36-18GHz are more reliable than that of high frequency, while over the shallow snow especially <15cm), the pair 36-18 is insensitive, but the high frequency pair (89/85-18) shows its distinct response, although the atmosphere influence is not be in consideration.Over Westrern (shallow snow situation), encourage using the emission information at high frequency when snow depth (< 20cm) , and the atmosphere influence is necessary counted in the later quantitative calculation.
Thank you very much!Thank you very much!