Comparison of Three Statistical Downscaling Methods and Ensemble Downscaling Method Based on Bayesian Model Averaging in Upper Hanjiang River Basin, China

Table 2

Results of three statistical downscaling methods (calibration and validation).

Statistical indicators	Season	Calibration				Validation
		Observation	SVM	BCC/RCG-WG	SDSM	Observation	SVM	BCC/RCG-WG	SDSM
Mean/mm	Spring	2.30	2.17	2.04	2.12	1.76	1.61	1.54	1.83
	Summer	4.33	4.41	4.19	4.37	4.03	4.06	3.75	4.19
	Autumn	2.95	2.78	2.70	2.83	2.01	2.12	1.83	2.05
	Winter	0.37	0.42	0.46	0.39	0.41	0.45	0.51	0.43
	Annual	2.49	2.46	2.23	2.38	2.06	2.27	2.32	2.15
P_95q/mm	Spring	22.96	18.72	17.48	20.92	22.85	18.22	17.71	18.91
	Summer	41.98	35.51	30.84	36.2	41.25	31.56	30.69	34.69
	Autumn	26.98	25.41	20.56	23.96	24.88	22.28	20.3	21.14
	Winter	7.58	5.74	4.97	6.13	7.71	5.47	4.88	5.53
	Annual	28.51	23.99	20.93	25.65	28.85	23.06	21.13	23.83
P_M5/mm	Spring	114.16	122.68	111.79	126.21	98.83	93.37	90.92	99.56
	Summer	233.99	190.81	165.14	207.7	213.71	181.54	180.13	181.07
	Autumn	171.39	178.52	134.62	155.47	126.44	123.59	109.81	124.24
	Winter	36.09	32.14	28.98	31.32	33.15	23.86	24.09	22.9
	Annual	236.46	216.87	171.33	213.85	214.35	186.23	186.82	185.66
CDD/d	Spring	23	20	18	19	20	21	15	24
	Summer	19	17	13	18	23	24	11	20
	Autumn	25	23	23	22	26	29	17	26
	Winter	49	37	33	38	54	50	25	40
	Annual	47	36	31	36	53	41	26	38
CWD/d	Spring	10	11	15	11	9	8	15	8
	Summer	12	11	17	13	11	9	18	10
	Autumn	17	16	19	17	15	12	20	10
	Winter	10	9	14	7	7	6	13	6
	Annual	17	19	20	18	15	13	23	11
R90t/%	Spring	39	36	44	37	36	40	44	38
	Summer	63	59	65	60	64	61	66	62
	Autumn	46	47	51	44	42	39	51	43
	Winter	6	7	5	4	5	4	5	3
	Annual	51	52	54	48	50	50	54	50