Interpretation of genotype-environment-sowing date/plant density interaction in sorghum [ Sorghum bicolor (L.) Moench] in early mature regions of China

Abstract

Sorghum [Sorghum bicolor (L.) Moench] is an important crop for food security in semiarid and arid regions due to its high tolerance to abiotic and biotic stresses and its good performance in marginal lands with relatively low fertility. To deeply understand the interrelationship among sorghum genotype, environment, sowing dates, and densities in the spring sowing early maturing (SSEM) areas of China, and to provide a basis for specifying scientific and reasonable cultural practices, a two-year field experiment was conducted with six popular varieties at six locations. Combined ANOVA showed that the yield difference between years was significant (P<0.05); the yield differences among locations, varieties, sowing dates, and densities were all highly significant (P<0.01). The variety effect was mainly influenced by location, year, sowing dates and their interactions. The sowing effect was mainly influenced by the location, year, variety and their interactions The plant density effect was significantly influenced by location and location-year interaction. Of the contributions of various test factors to yield variance, the location was the largest one (38.18%), followed by variety (12.31%), sowing date (1.53%), density (0.54%), and year (0.09%), with all these single factors accounting for 52.65%. The total contribution of all two-factor interactions accounted for 14.24%, among which the greatest contributor was location-hybrid interaction (8.07%). The total contribution of all three-factor interactions accounted for 14.58%, of which year-location-hybrid interaction was the largest contributor (9.02%). Sowing dates significantly affected model of sorghum growth and development, especially during the late period. The key climatic factors affecting yield were different among the six locations. Weather factors during the grain filling stages contributed much more than those during the early stage to grain yield. Mid-maturing varieties are recommended other than early maturing varieties for the SSEM areas even when late sowing occurs. Sowing as early as possible is recommended for areas with very short frost-free period (Harbin, Tongliao, and Datong). Proper delayed sowing is recommended for areas with a relative long frost-free period (Gongzhuling, Baicheng and Zhangjiakou). This research will provide a conducive reference for sorghum production in similar areas.

Keywords: Sorghum bicolor; environment; genotype; plant density; sowing date.

Figure 1

Locations of the study stations. HH, Harbin of Heilongjiang; JB, Baicheng of Jilin; JG, Gongzhuling of Jilin; IT, Tongliao of inner Mongolia; HZ, Zhangjiakou of Hebei; SD, Datong of Shanxi.

Figure 2

General yield performance at six locations in 2020-2021. a, b, c, d, c, e, (A–D), indicating mean yield difference significance among locations at 0.05, and 0.01; *, ** and *** indicating difference significance between two years at 0.05, 0.01 and 0.001.

Figure 3

Relationship among locations under different sowing dates revealed by RDA analysis.

Figure 4

Relationship of location and ecological factors revealed by RDA analysis of yield. EAT, effective accumulative temperature; AAT, accumulative average temperature; AGT5, average 5-cm ground temperature; AGT10, average 10-cm ground temperature; ARH, average relative humidity per day; TRF, total rainfall; TSH, total sunshine. -1, stage from emergence to flowering; -2, stage from flowering to mature.

Figure 5

Yield effects of different sowing period and densities. (A) Yield of different sowing dates under two densities; (B) Interaction of sowing dates and density. Letters a and b are used to label the difference significance at 0.05; the same letter means not significant (NS) among the three sowing dates; ** means highly significant at 0.01 between the two densities.

Figure 6

RDA analysis of yields on sowing dates in response to climatic factors during emergence to flowering and flowering to maturity in 2020 (left) and 2021(right). EAT, effective accumulative temperature; AAT, accumulative average temperature; AGT5, average 5-cm ground temperature;AGT10, average 10-cm ground temperature; ARH, average relative humidity per day; TRF, total rainfall; TSH, total sunshine. -1, stage from emergence to flowering; -2, stage from flowering to mature.

Figure 7

“Which Won Where/What” biplot under different sowing dates.

Figure 8

Interactions between sowing dates and variety (left), years (mid), and locations (right). Letters a, b and c are used to label the difference significance at 0.05; the same letter means not significant (NS), and a different letter means significant among the three sowing dates.

Figure 9

Effect of sowing date on sorghum grain yield in different locations (Letters a–c), after data means difference significance at 0.05, the same letters mean not significant).

Figure 10

Effects of sowing dates on sorghum development progress in different locations in 2020-2021.Letters (a–c) are used to label the difference significance at 0.05, and the same letter means not significant.

Figure 11

“Which Won Where/What” biplot under different sowing dates (Left, early sowing, S1; Right, late sowing, S3).