Impact of high temperature on heterosis and general combining ability in spring canola (Brassica napus L.)

Abstract

Canola is an extremely important commodity to the western Canadian economy. Annual or geographic fluctuations in yield due to heat stress can have a significant influence on total production. To increase the stability of canola yield, breeders need to select for increased heat tolerance in a reliable, repeatable, cost effective manner that represents commercial production conditions. Field experiments with 10 (5 maintainer and 5 restorer) inbreds and the 25 corresponding hybrids were conducted across 2 years with 2 locations per year. Each experiment had two planting dates (normal planting date and a late planting date). The experiments were designed to determine if: 1) planting date can be used as a proxy for heat stress; 2) heterosis can minimize the impact of heat stress; and 3) direct selection within a heat-stress environment improves genetic gain for heat-stress tolerance. Principal component analysis of the Tmax, Tmin, Tmean, soil moisture, daily light integral and yield were conducted. The resulting biplot analyses demonstrate that the temperature variables had a greater impact on yield than soil moisture or daily light integral and therefore planting date was identified as a successful proxy for heat stress. The heat-stress treatment caused a 23% mean reduction in seed yield (20% reduction within the hybrids and 25% reduction within the inbreds) indicating that heterosis can reduce the impact of higher temperatures on canola yield. The mean heat susceptibility index was 0.94 and 1.20 for the hybrids and inbreds respectively; further demonstrating that the hybrids were more heat tolerant than the inbreds. Analyses of the GCA for the inbreds shows 4 cross-over interactions between the two temperature treatments indicating that direct selection for heat tolerance will be necessary to increase the response to selection for this trait. Overall canola breeders interested in increasing heat tolerance need to consider utilizing planting date as a high-temperature treatment and cannot rely on indirect selection from experiments under optimal conditions.