Validation of TRMM rainfall products at Gadanki T. Narayana Rao NARL, Gadanki K. Nakamura, HyARC, Nagoya, Japan D. Narayana Rao, NARL, Gadanki National Atmospheric Research Laboratory (NARL), Gadanki
Jan 20, 2016
Validation of TRMM rainfall products at Gadanki
T. Narayana RaoNARL, Gadanki
K. Nakamura, HyARC, Nagoya, Japan D. Narayana Rao, NARL, Gadanki
National Atmospheric Research Laboratory (NARL), Gadanki
2 4 6 8 10 12
M onths
Tim
e (L
T)
0
50
100
150
0
2
4
6
8
10
0
500
1000
1500
2000
2500
21 - 24
15 - 18
09 - 12
03 - 06
(a) N um ber o f ra in occurrences
(b) M ean ra in rate
(c) Accum ulated ra in
21 - 24
15 - 18
09 - 12
03 - 06
21 - 24
15 - 18
09 - 12
03 - 06
O RGDiurnal variation
National Atmospheric Research Laboratory (NARL), Gadanki
2
4
6
8
Hei
ght (
km)
-15
-5
5
15
25
35
Tim e (LT)00 00
0
2
4
6
8
10
10 20 30 40
2 :00 LT 2:11 LT 2:22 LT
dBZe
13.48 N , 79.22 E 13.53 N , 79.20 E 13.50 N , 79.18 E 13.47 N , 79.17 E 13.53 N , 79.15 E 13.50 N , 79.13 E
Comparison of vertical profile of Ze
National Atmospheric Research Laboratory (NARL), Gadanki
2 4 6 8 10 12
M onths
2
4
6
8
He
ight
(km
)
50
250
450
650
850
2
4
6
8
Hei
ght (
km)
0
30
60
90
120(a) LAW P
(b) TR M M
C olour scale is for to ta l num ber of occurrencesM ean values are ind icated by " + "
Jan Apr Jul O ct3.5
4.0
4.5
5.0
5.5 0 C isotherm level LAW P TR M M
M onth
Bright band statistics
National Atmospheric Research Laboratory (NARL), Gadanki
• Significant diurnal variation of rainfall is seen in SW monsoon season, whereas in the NE monsoon season it is negligible
• TRMM reproduces the diurnal variation of precipitation seen with the ORG
• Vertical profile of Ze given by profiler agrees well with that of TRMM below the bright band. The enhancement of Ze at BB is more in the case of profiler. Though, one can expect different Ze for different radars operating at different frequencies at high Z values (or at high rain rates), the difference seems to be more than one can expect.
• The bright band height variation during the course of the annul cycle is less than 250 m
• Bright band height is ~ 500 m below the 0 C isotherm level
• Height of the bright band as seen by TRMM matches with that of profiler
Conclusions
National Atmospheric Research Laboratory (NARL), Gadanki
… Thank you
Selection of suitable cases for radar calibration24 Feb 00
R eflectiv ity factor (dBZe)
D oppler ve locity (m /s)
Spectra l w idth (m /s)
Surface observations
2468
0
10
20
30
2468
-8
-6
-4
-2
0
2
2468
Hei
ght (
km)
0
0.5
1
1.5
2
4 6 8 10 12 14 16 18
20
30
40 Z
Tim e (LT)
0
4
8
Rdisdro
Rorg
2468
0
10
20
30
2468
-8
-6
-4
-2
0
2
2468
Hei
ght (
km)
0
0.5
1
1.5
2
19 N ov 99 R eflectiv ity factor (dBZe)
D oppler ve locity (m /s)
Spectra l w idth (m /s)
Surface observations
4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4
2 0
3 0
4 0
5 0 Z
T i m e ( L T )
0
4
8
1 2
1 6
2 0
Rd i s d r o
Ro r g
Profiler calibration
Bright band detection with a wind profiler
The algorithm looks for a peak in vertical profiles of Z in the height range of 3.5-5.5 km
+
A Doppler velocity difference of 1.5 m/s and a spectral width change of 0.3 m/s in 900 m (± 3 range bins from the Z peak).
Several rain events are studied in detail to arrive at these thresholds.
BB detection in case studies - I
28nov99R eflectiv ity factor (dBZe)
D oppler ve locity (m /s)
Spectra l w idth (m /s)
Surface observations
2
4
6
8
0
10
20
30
40
2
4
6
8
-8
-6
-4
-2
0
2
2
4
6
8
Hei
ght
(km
)
0
1
2
6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4
2 0
3 0
4 0
5 0 Z
T i m e ( L T )
0
2 0
4 0
6 0
Rd i s d r o
Ro r g
5 O ct 99Reflectiv ity factor (dBZe)
Doppler velocity (m /s)
Spectra l w idth (m /s)
Surface observations
2
4
6
8
0
10
20
30
40
2
4
6
8
Hei
ght
(km
)
0
1
2
2
4
6
8
-8
-6
-4
-2
0
2
0 2 4 6 8 10 12 14 16 18 20 22 24
20
30
40
50 Z
Tim e (LT)
0
4
8
12
16
Rdisdro
Rorg
BB detection in case studies - II22 - 23 June 00
R eflectiv ity factor (dBZe)
D oppler ve locity (m /s)
Spectra l w idth (m /s)
Surface observations
2
4
6
8
0
10
20
30
40
50
2
4
6
8
-8
-6
-4
-2
0
2
2
4
6
8
Hei
ght (
km)
0
1
2
2
4
6
8
0 6 1 2 1 8 2 4 3 0 3 6
2 0
3 0
4 0
5 0 Z
T i m e ( L T )
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
Rd i s d r o
Ro r g
28 - 29 June 00
R eflectiv ity factor (dBZe)
D oppler velocity (m /s)
Spectra l w idth (m /s)
Surface observations
2
4
6
8
0
10
20
30
40
50
2
4
6
8
-8
-6
-4
-2
0
2
2
4
6
8
Hei
ght (
km)
0
1
2
0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0
2 0
3 0
4 0 Z
T i m e ( L T )
0
2
4
6
8
1 0
Rd i s d r o
Ro r g
Data
• TRMM – 2A25 5 years (1998-2002)
• 1.3 GHz Wind profiler Precipitation in 3 rainy seasons
• Disdrometer 5 years
• ORG 5 years
Errors: Random & Measurement errors, Systematic biases ??
Resolution: 1° x 1° and 2° x 2° , near nadir and full FOV
For comparison : Surface rain (rainfall, rainrate, rain occurrence), diurnal variation of rainfall, vertical profiles and bright band statistics.
Location
2 4 6 8 10 120
50
100
150
200
M onth
G adanki Tam ilnadu S . Int. Karnataka Rayalaseem a Coasta l Andhra
The comparison of rain parameters at the proposed site has its significance both technically and geographically.
Technically – Profilers operating at VHF and UHF bands and other ground based precipitation measurements.
Geographically – Gadanki is located close to the south-east coast of India, where the rainfall peaks in the northeast monsoon, in contrast to the rainfall in rest of India