Low temperature during reproductive development (panicle initiation to maturity) is one of the major factors limiting productivity in the NSW rice industry. Low temperatures during late January to early February disturb normal development of the pollen grains, causing spikelet sterility. The risk of yield reduction in a low temperature year is greatly enhanced by increased nitrogen status of the crop (Heenan 1984, Satake et al. 1987). Hayase et al. (1969) concluded that the young microspore stage, which is related to male sterility, is the most sensitive stage to low temperatures for rice plants. Low temperatures during reproductive development reduce the number of engorged pollen grains and fertilised spikelets in rice (Ito 1971). Deep irrigation water (20cm) during the reproductive period can help protect the young panicles from low air temperatures. It provides a buffer and increases the panicle temperature by up to 7oC on a cool night (Williams and Angus 1994).
A recent glasshouse experiment at the University of Queensland found a significant positive correlation between total number of engorged pollen grains produced in an anther and the number of pollen grains intercepted by the stigma (r2= 0.81). This correlation suggests that 600 engorged pollen grains per anther would result in greater than 40 intercepted on the stigma (Figure 4-A). There was a significant negative correlation between number of engorged pollen grains and spikelet sterility (r2= 0.59). The relationship suggests that 600 engorged pollen grains will result in less than 30% sterility (Figure 4-B). These correlations indicate that a large number of engorged pollen per anther is the key to successful fertilisation.
The critical temperature for inducing sterility varies amongst cultivars. The unpredictability of low temperature during the microspore stage of rice has caused severe yield losses throughout the world. Low temperatures throughout Japan in 1993 led to the opening up of their markets for the importation of rice. An extended low temperature event in Australia during the reproductive stage in 1996 reduced yields across the rice industry by 25%. In Laos the 1999/2000 season was the coolest since 1974, causing major shortfalls in rice production. Satake (1969) estimated the critical temperatures were 15°C to 17oC in a tolerant cultivar and 17°C to 19oC in a susceptible one. Australian rice crops are exposed to reproductive low temperature damage from late-January through to early-February. Long term data show that this period is usually the warmest with an average minimum temperature of 17oC. However, in the dry season crops in Laos the time of reproductive development is more variable, occurring from early-March through to mid-April. The average minimum temperature across five Laos provinces increased by 3.5oC from early March to early April (Figure 5). The average minimum temperature in Oudonxay in Northern Laos was increased from 12oC in early March to 16.4oC in early April. Delaying reproductive development by sowing later may reduce the extent of low temperature damage in Laos.
A series of two experiments investigating genotypic variation for low temperature tolerance (36 and 18 cultivars respectively) was recently completed in temperature controlled facilities at Yanco Agricultural Institute. Three day/night temperature regimes (32/25, 25/15 and 27/13oC) were imposed on cultivars from after panicle initiation through to head emergence. A combined analysis identified seven international cultivars that consistently performed better than all the Australian cultivars. The cultivars include Liman and Pavlovsky (from Russia), Plovdiv 22 (Bulgaria), Akihikari and Haenuki (Japan), HSC55 (Hungary) and M103 (California). Low temperatures reduced harvest index (grain/total biomass) of these tolerant cultivars by only 20% compared to 50% for the major Australian cultivars.
Field trials at Yanco Agricultural Institute during the 1998/99 and 1999/2000 seasons aimed to confirm the tolerance of cultivars in the field. The 1998/99 season consisted of nine sowing dates from early October to late December with each sowing date including a replicated trial of 30 genotypes. However attempts to confirm cold tolerance in the field was thwarted by the occurrence of above average temperatures. The 1999/2000 field trial comprised six sowing dates from October 5 to December 30, 1999. Deep (22 cm) and shallow (5 cm) water depth treatments were imposed through the critical young microspore stage. Twenty-eight cultivars from different origins, with varying susceptibility to low temperatures, were replicated in each of the 12 bays. Low night temperatures during the late December-early January and late January periods caused significant levels of sterility in most cultivars in the shallow water treatments. Liman (Russia), M103 (America) and Hitomebore (Japan) had a low level of sterility despite experiencing low temperatures during critical stages in late January. Sprint (Russia), Doongara (Australia) and Leng Kwang (China) had high levels of sterility in the shallow water treatments and appear to be susceptible to mid-season low temperatures.
