Abstract:

To analyze the heat tolerance of japonica rice in Liaohe plain rice-growing region, and establish an effective evaluation system of heat tolerance, 20 main varieties of japonica rice suitable for cultivating over this region were selected and planted under simulated temperature increase using the free air temperature increase system. The heat tolerance of different japonica rice varieties was comprehensively evaluated by multivariate statistical analysis methods based on the heat tolerance coefficients of 10 rice trait indicators, the effective panicle number (X₁), grain number per spike (X₂), seed setting rate (X₃), 1000-grain weight (X₄), yield (X₅), brown rice rate (X₆), milled rice rate (X₇), protein content (X₈), amylose content (X₉), and taste value (X₁₀) after harvesting and air-drying the rice. The results showed that the trait indicators except taste value were significantly or extremely reduced under the warming. Correlation analysis based on heat tolerance coefficients revealed different degrees of association between rice trait indicators. Principal component analysis condensed the 10 individual indicators into 4 mutually independent comprehensive indicators. The 4 comprehensive indicators had contribution rates of 34.816%, 19.265%, 15.636%, and 10.605%, respectively, giving a cumulative contribution rate of 80.322%. The comprehensive evaluation values of heat tolerance (D_i value) were calculated using the membership function method. The higher the D_i value, the stronger the heat tolerance. The heat tolerance of 'Yanjing 456' was the strongest and that of 'Shennong 016' was the worst. The 20 japonica rice varieties were classified into three categories based on their heat tolerance: the first category comprised 7 heat-tolerant varieties, the second category included 6 moderately heat-tolerant varieties, and the third category consisted of 7 heat-sensitive varieties.

Key words: Japonica rice, Warming, Heat tolerance, Comprehensive evaluation, Multivariate statistical analysis