Abstract Optimizing grain production implies defining the best management practices for a set of particular environments. Argentinean farmers in the central temperate region are sowing maize at two contrasting sowing dates (September to October and December), exposing their crops to very different growing environments. We tested the influence of management and environmental variables affecting maize yield at early (ES) or late (LS) sowings. Our objectives were to (i) determine the most relevant management and environmental yield predictors for ES and LS, (ii) quantify the magnitude of their effects, and (iii) explore potential yield increments after optimizing crop management within each sowing. We conducted 91 on-farm multi-environment trials during six years (2010–2016) around the central temperate region, and analyzed 13 management and environmental variables. The most relevant environmental predictors (relative importance > 0.50) for both sowing dates included presence of a water table at sowing, rainfall during the crop cycle, and their interactions. Presence of a water table had a positive or negative effect for ES or LS, respectively. Management yield predictors varied depending on the sowing date. Stand density, N and S availability were important yield predictors at ES, while fungicide use, soil P, and N availability were the most relevant ones at LS. Farmers can increase yield at each sowing date by optimizing these management practices. Optimizing stand density and N availability in ES can have a ∼3,053 kg ha−1 effect, while fungicide use in LS can increase yield by ∼1040 kg ha−1. Determining the adequate sowing date based on the presence of a water table at sowing can have a ∼1000 kg ha−1 effect. Our results described specific management options for reducing yield gaps and optimize maize production across contrasting sowing dates.

中文翻译：

---

减少不同播种日期中玉米产量差距的管理方案

摘要 优化粮食生产意味着为一组特定环境定义最佳管理实践。中温带地区的阿根廷农民在两个截然不同的播种日期（9 月至 10 月和 12 月）播种玉米，使他们的作物暴露在截然不同的生长环境中。我们测试了管理和环境变量在早期 (ES) 或晚期 (LS) 播种时影响玉米产量的影响。我们的目标是 (i) 确定 ES 和 LS 最相关的管理和环境产量预测因子，(ii) 量化其影响的大小，以及 (iii) 在优化每次播种的作物管理后探索潜在的产量增量。我们在 6 年（2010-2016 年）中围绕中温带地区进行了 91 次农场多环境试验，并分析了 13 个管理和环境变量。两个播种日期最相关的环境预测因子（相对重要性 > 0.50）包括播种时地下水位的存在、作物周期中的降雨量及其相互作用。地下水位的存在分别对 ES 或 LS 产生正面或负面影响。管理产量预测因子因播种日期而异。林分密度、N 和 S 可用性是 ES 的重要产量预测因子，而杀菌剂使用、土壤 P 和 N 可用性是 LS 最相关的。农民可以通过优化这些管理实践来提高每个播种日期的产量。优化 ES 中的林分密度和 N 可用性可以产生 ~3,053 kg ha-1 的效果，而 LS 中使用杀菌剂可以增加 ~1040 kg ha-1 的产量。根据播种时地下水位的存在来确定适当的播种日期可能会产生约 1000 kg ha-1 的影响。我们的结果描述了在不同播种日期之间减少产量差距和优化玉米生产的特定管理选项。