Table 4.

Table showing the predictive ability for 5 traits in four datasets

Locations	Traits	ST-CV1	MT-CV1	MT-CV2	% Increase from ST-CV1 to MT-CV2
BLUEQ17	HI	0.27	0.29	0.32	18.5
	GY	0.18	0.17	0.35	94.4
	GN	0.21	0.20	0.50	138.1
	SPI	0.11	0.11	0.18	63.6
	FE	0.07	0.07	0.09	28.6
BLUEQ18	HI	0.39	0.40	0.41	5.1
	GY	0.22	0.21	0.41	86.4
	GN	0.23	0.22	0.42	82.6
	SPI	0.22	0.22	0.26	18.2
	FE	0.21	0.19	0.22	4.8
BLUEC18	HI	0.31	0.30	0.42	35.5
	GY	0.21	0.23	0.50	138.1
	GN	0.13	0.13	0.31	138.5
	SPI	0.18	0.20	0.25	38.9
	FE	0.13	0.14	0.15	15.4
BLUEAll	HI	0.31	0.32	0.46	48.4
	GY	0.20	0.21	0.39	95.0
	GN	0.14	0.16	0.33	135.7
	SPI	0.16	0.17	0.17	6.3
	FE	0.17	0.17	0.19	11.8

Single-trait prediction model (ST-CV1), and multi-trait prediction mode (MT) with two schemes of cross-validation (MT-CV1 and MT-CV2); HI harvest index, GY grain yield in kg ha⁻¹, GN grain number m⁻², SPI spike partitioning index, FE fruiting efficiency in grains g^− 1 of spike dry weight at anthesis+ 7 days, NDVI normalized difference vegetation index, CT canopy temperature in ^oC