Rainfall screening methodology using TRMM data over a river basin

Figures & data

Table 1. Global rain/no-rain classification algorithms based on scattering index, SI.

Proposer	Observed Tb	Tb,est	Threshold of SI
Grody (1991)	85 V		SI > 10
Adler et al. (1994)	85 H	251	SI > 4
Kummerow and Giglio (1994)	85 H	min[Tb,37H, 265]	SI > 0
Ferraro (1997)	85 V		SI > 10
Kummerow (2001) (GPROF)	85 V	Tb,22V	SI > 8

Figure 1. Land use/land cover in the Mahanadi basin (2010).

Figure 2. Topography of the Mahanadi basin from the GTOPO digital elevation model. ▴, locations of the automatic weather stations within the basin.

Table 2. Automatic weather stations (AWS) located within the Mahanadi basin.

AWS	Latitude	Longitude
Raighar	19.88	82.07
Chilka Khurda	19.67	85.18
Bolagarh	20.16	85.27
Gop	19.97	86.01
Pipli	20.10	85.98
Ersama	20.20	86.40
Garadapur	20.38	86.37
Kendrapara	20.42	86.57
Kendrapara	20.49	86.43
Chowdar	20.56	85.99
Badachana	20.67	86.11
Bari	20.67	86.35
Sinapali	20.43	82.63
Komna	20.50	82.67
Patnagarh	20.71	83.12
Rengali	21.64	84.03
Bamara	22.05	84.28
Golamunda	20.05	82.78
Khariar	20.30	82.75
Sinapali	20.43	82.63
Tureikela	20.50	82.80
Khaprakhol	20.80	82.86
Gudvela	20.41	83.56
Agalpur	20.97	83.49
Boudh	20.84	84.28
Redhakhol	21.07	84.33
Somenahalli	21.36	83.74
Ambabhona	21.58	83.74
Sambalpur	21.41	84.03
Junagarh	19.86	82.94
Bhawanipatna	19.91	83.16
M. Rampur	20.19	83.51
Nuagaon	20.13	84.01
Raikia	20.05	84.23

Figure 3. Flowchart summarizing the methodology adopted.

Figure 4. Relationship of near surface rain rate (NSR) with collocated over-land T_b (from TMI) for the Mahanadi basin.

Table 3. Contingency matrix.

	Rain judged by PR	No rain judged by PR
Rain judged by TMI	a	b
No rain judged by TMI	c	d

Table 4. Performance statistics used in the study.

Performance measure	Formula	Range
Probability of detection (POD)		[0, 1]
Threat score (TS) or critical success index (CSI)		[0, 1]
False alarm ratio (FAR)		[0, 1]
Heidke skill score (HSS)		[−∞, 1]
Kuiper skill score (KSS)		[−1, 1]
Percentage correct (PC) or hit rate (H)		[0, 1]
Bias (B)		[0, ∞]
Gilbert skill score (GSS)	,	[0, 1]
Odds ratio skill score (ORSS)		[−1, 1]
Log odds ratio		[−∞, ∞]

Figure 5. Performance diagram showing categorical statistics for the 15 days. The triangles represent each of the 15 case study dates. Their locations on the performance diagram give the values of performance statistics (POD, B, FAR, CSI), with cross hairs representing the confidence interval obtained using 1000 bootstrapped samples from the contingency table. Success ratio = 1 − FAR.

Table 5. Results of dichotomous classification using proposed model for the 15 days.

Date	No. of orbits	POD	FAR	TS	GSS	B	ORSS	KSS	HSS	PC	Log theta
13 June 2011	2	0.88	0.24	0.68	0.66	1.17	0.99	0.85	0.79	0.96	5.74
17 June 2011	3	0.83	0.18	0.70	0.60	1.01	0.96	0.75	0.75	0.90	4.16
25 June 2011	4	0.78	0.22	0.63	0.55	1.01	0.96	0.72	0.71	0.90	3.99
2 July 2011	2	0.60	0.30	0.48	0.43	0.86	0.95	0.57	0.60	0.92	3.83
7 July 2011	4	0.89	0.12	0.78	0.70	1.02	0.98	0.83	0.82	0.92	4.90
12 July 2011	2	0.70	0.31	0.53	0.47	1.02	0.95	0.64	0.64	0.90	3.68
8 August 2011	3	0.86	0.36	0.58	0.48	1.35	0.95	0.73	0.65	0.87	3.77
17 August 2011	4	0.85	0.35	0.58	0.52	1.32	0.97	0.77	0.68	0.90	4.21
25 August 2011	4	0.91	0.21	0.73	0.58	1.16	0.96	0.76	0.73	0.87	4.05
1 September 2011	4	0.84	0.34	0.58	0.43	1.27	0.91	0.65	0.60	0.82	3.14
18 September 2011	2	0.85	0.25	0.66	0.57	1.14	0.96	0.77	0.73	0.90	4.15
23 September 2011	4	0.80	0.41	0.51	0.49	1.38	0.98	0.78	0.66	0.96	4.98
21 June 2012	3	0.90	0.32	0.63	0.44	1.34	0.93	0.66	0.62	0.81	3.42
30 July 2012	3	0.84	0.17	0.72	0.59	1.02	0.96	0.75	0.74	0.88	3.97
24 August 2012	1	0.89	0.09	0.81	0.71	0.98	0.98	0.82	0.83	0.91	4.81

Table 6. Results of dichotomous classification based on rainfall type for the 15 days.

Date	Convective					Stratiform
	No. of footprints	Average rain rate (mm/h)	Percentage convective	POD	FAR	No. of footprints	Average rain rate (mm/h)	Percentage stratiform	POD	FAR
13 June 2011	242	14.60	27.40%	0.97	0	641	1.63	72.59%	0.85	0.29
17 June 2011	488	10.88	21.09%	0.92	0	1825	1.91	78.90%	0.81	0.22
25 June 2011	200	13.67	14.62%	0.88	0	1162	1.51	85.31%	0.77	0.25
2 July 2011	182	10.53	22.47%	0.74	0	628	1.67	77.53%	0.57	0.35
7 July 2011	157	9.67	15.24%	0.90	0	873	1.82	84.75%	0.89	0.14
12 July 2011	294	14.16	27.73%	0.80	0	766	1.56	72.26%	0.67	0.38
8 August 2011	224	13.23	14.24%	0.87	0	1348	1.31	85.75%	0.86	0.39
17 August 2011	28	12.21	5.45%	0.89	0	485	1.02	94.54%	0.85	0.36
25 August 2011	439	11.41	17.45%	0.96	0	2076	1.71	82.54%	0.90	0.24
1 September 2011	128	12.20	12.61%	0.95	0	887	1.57	87.38%	0.82	0.37
18 September 2011	457	13.48	19.94%	0.91	0	1834	1.68	80.05%	0.84	0.29
23 September 2011	49	10.1821	18.49%	0.9183	0	192	1.8704	81.50%	0.784	0.47
21 June 2012	197	10.8596	7.87%	1	0	2306	1.3464	92.12%	0.8991	0.34
30 July 2012	739	10.8988	23.72%	0.9539	0	2376	2.0332	76.27%	0.8163	0.21
24 August 2012	118	9.53	17.48%	0.96	0	557	1.90	82.51%	0.87	0.11

Figure 6. Performance diagram showing categorical statistics for stratiform rainfall (as explained for Figure 5).

Figure 7. Accumulated rainfall over the Mahanadi basin during JJAS 2011 showing (a) convective rainfall and (b) stratiform rainfall.

Figure 8. TRMM 3B42 product showing accumulated rainfall (mm) for the Mahanadi basin on (a) 2 July 2011 and (b) 21 July 2011.

Table 7. Results of proposed RNC algorithm validated with AWS raingauges and PR rainflags.

Date	POD		FAR		TS		B
	PR	AWS	PR	AWS	PR	AWS	PR	AWS
7 July 2011	0.89	0.75	0.12	0.33	0.78	0.54	1.02	1.12
30 July 2012	0.84	0.75	0.17	0.39	0.72	0.49	1.06	1.25
17 June 2011	0.83	0.55	0.18	0.37	0.60	0.41	1.01	0.88
23 September 2011	0.80	0.12	0.41	0	0.51	0.12	1.38	0.125

Grody, N.C., 1991. Classification of snow cover and precipitation using the special sensor microwave imager. Journal of Geophysical Research, 96, 7423–7435. doi:10.1029/91JD00045

 Web of Science ®Google Scholar

Kummerow, C. and Giglio, L., 1994. A passive microwave technique for estimating rainfall and vertical structure information from space. Part I: algorithm description. Journal of Applied Meteorology, 33, 3–18. doi:10.1175/1520-0450(1994)033<0003:APMTFE>2.0.CO;2

 Google Scholar

Ferraro, R.R., 1997. Special Sensor Microwave Imager derived global rainfall estimates for climatological applications. Journal of Geophysical Research, 102, 16715–16735. doi:10.1029/97JD01210

 Web of Science ®Google Scholar

Kummerow, C., 2001. The evolution of the Goddard profiling algorithm (GPROF) for rainfall estimation from passive microwave sensors. Journal of Applied Meteorology, 40, 1801–1820. doi:10.1175/1520-0450(2001)040<1801:TEOTGP>2.0.CO;2

 Google Scholar