我国北部冬麦区小麦区域试验重复次数和试点数量的优化设计

农作物品种区域试验重复次数和试点数量的合理配置有利于提高试验的成本效率和品种选择效率。本研究分析了2010—2019年期间北部冬麦区小麦品种区域试验的重复次数和试点数量设置的合理性,依据小麦品种试验的信噪比和遗传力水平随重复次数和试点数量的变化规律,提出了重复次数和试点数量的优化设计方案。结果表明:(1)北部冬麦区小麦单点试验的遗传力平均达到0.87,需要的重复次数平均值仅为1.4,说明3次重复可以充分保证试验精确度的需求。(2)北部冬麦区水地组和旱地组小麦区域试验达到0.75的遗传力水平时,需要的试点数量分别为11个和13个,目前有效试点数量分别约为11个和8个,分别达到0.75和0.60的遗传力水平。(3)小麦品种区域试验结果对品种的审定和应用十分重要,而每年都可能有少数试验点因为各种异常情况而报废,为保证试验结果的可靠性,可按H=0.75的水平需求安排试验点数量和重复次数,即重复次数可保持当前的3次;水地组的试点数量可保持在11个左右;旱地组可将试点增加到13个;如要将遗传力提高到0.80的水平,则需约16个试点。

Optimization of test location number and replicate frequency in regional winter wheat variety trials in northern winter wheat region in China

The rational allocation of the replicate frequency and test location number in regional crop trials are highly beneficial to the enhancement of both trial cost efficiency and new cultivar selection efficiency. The rationality of test location number and replicate frequency for national wheat regional trials in the Northern winter wheat region (NWWR) in China was evaluated by using historical experimental datasets during the last 10 years according to the dynamics of trial heritability and noise-signal quotient (Q) with the increase of test locations and replicates within trials in 2010-2019, proposing an optimal design scheme of the replicate frequency and test location number for the wheat planting region. The result indicated three replicates at current achieve 0.87 of the averaged within-trial heritability of single-trials in NWWR, and only 1.4 times of replicates on average was needed to achieve 0.75 of within-trial heritability, so three replicates were obviously sufficient to maintain enough test accuracy. The needed test location number to achieve 0.75 of cross-trial heritability was estimated as 11 for the irrigated group and 13 for the rainfed group in the one-year multi-locational trials in NWWR, while in the current wheat trials that was about 11 and 8 effective test locations, achieving a heritability level of 0.75 and 0.60, respectively. The irrigated group was exactly meet the trial accuracy requirement (H = 0.75), while the rainfed group was somewhat insufficient. Considering the importance of the regional wheat trials in recommending new varieties for registration and production, and the possible trial cancellation due to various abnormalities, the practical management should set a heritability level around 0.75, and maintain three replicates, with around 11 test locations for the irrigated group, and 13 locations for the rain-fed group. In order to increase heritability to the level of 0.80, about 16 test locations should be required.