Soybeans are of great importance to the global economy, but the cultivation as monoculture has shown several negative environmental implications in the long-term. Long-term studies demonstrate the cumulative effects of rotations on soil variables, but few studies have considered changes during consecutive years in a time series of soybean as a monoculture. The inclusion of cereals and cover crops in rotation with soybean increases the intensification sequence index (ISI, time with crops actively growing during the year) and increases in soybean yield are expected. In line with this, some studies suggest that only including a winter cover crop will increase ISI and raise production. However, these comparisons have not been made in the long term. The objectives of this work were to: i) compare the evolution of full-season soybean yield and production when sowed as monoculture and in cropping sequences that include cereal crops and cover crops, and ii) quantify a yield gap in full-season soybean due to monoculture by evaluating the long-term residual effect of crop sequences with different land occupation. Soybean yield time series in a 14-years period in rotation or as monoculture were studied in a long-term field experiment under no-tillage established in 2006 in the Northern Pampas region of Argentina. Rotations consisted in sequences that included soybean (S), maize (M) and wheat (W), and also incorporated wheat as a winter cover crop (CC): S-S, CC/S-CC/S, S-W/S-M, CC/S-W/S-M, M-W/S, and CC/M-W/S, with ISI values of 0.39, 0.69, 0.55, 0.64, 0.65 and 0.80, respectively. In 2021/22 season, full-season soybean was planted in all plots as a "test" crop, to evaluate the long-term residual effect of sequences with different intensification indexes. Seed yield averaged 3249 kg ha−1 among years. During the 14 years under study, seed yield in S-W/S-M surpassed S-S by 28 % in 6 out of 14 years. The differences between S-S with respect to this sequence were observed consistently from the ninth crop season (i.e. since 2017/18). The inclusion of a cover crop within a soybean monoculture (i.e. CC/S-CC/S) showed similar seed yields as in S-S in 11 out of 14 years. The analysis of the test crop showed yield-gaps when soybeans were grown as monoculture (seed yield in rotations minus seed yield in S-S) of 9, 20, 27, 21 and 31 % for CC/S-CC/S, S-W/S-M, CC/S-W/S-M, M-W/S and CC/M-W/S, respectively. In summary, soybean in rotation with cereals showed an average increase of ca. 346 kg ha−1 and including a cover crop within a soybean monoculture did not increase seed yield at the levels observed when rotated with cereals. Finally, when estimating seed yield-gaps in soybean, it is important to set in which rotation soybean is placed, since a higher maximum yield is expected to happen when soybean follows a more intensified rotation.

中文翻译：

---

在长期试验中，与单一种植相比，与谷物和覆盖作物轮作相比，全季大豆的种子产量更高


大豆对全球经济非常重要，但从长远来看，单一种植已经显示出一些负面的环境影响。长期研究表明轮作对土壤变量的累积影响，但很少有研究将大豆作为单一栽培的时间序列中连续几年的变化考虑。将谷物和覆盖作物与大豆轮作会增加集约化序列指数（ISI，作物在一年中积极生长的时间），并有望提高大豆产量。与此一致，一些研究表明，仅包括冬季覆盖作物会增加 ISI 并提高产量。然而，这些比较并没有从长远来看进行。这项工作的目标是：i） 比较作为单一作物播种和包括谷类作物和覆盖作物在内的种植序列时全季大豆产量和产量的演变，以及 ii） 通过评估不同土地占用的作物序列的长期残留效应，量化由于单一种植而导致的全季大豆产量差距。在 2006 年在阿根廷北部潘帕斯草原地区建立的免耕长期田间试验中，研究了 14 年轮作或单一栽培期间的大豆产量时间序列。轮作包括大豆 （S）、玉米 （M） 和小麦 （W），并将小麦作为冬季覆盖作物 （CC）：S-S、CC/S-CC/S、S-W/S-M、CC/S-W/S-M、M-W/S 和 CC/M-W/S，ISI 值分别为 0.39、0.69、0.55、0.64、0.65 和 0.80。在 2021/22 年度，所有地块均种植全季大豆作为“测试”作物，以评估具有不同集约化指数的序列的长期残留效应。 种子产量每年平均为 3249 kg ha-1。在研究的 14 年中，SW/S-M 的种子产量在 14 年中有 6 年超过 SS 28%。从第九季（即自 2017/18 年以来）开始，就此序列观察到 S-S 之间的差异。在大豆单一栽培中加入覆盖作物（即 CC/S-CC/S）在 14 年中有 11 年显示出与 SS 相似的种子产量。对测试作物的分析显示，当大豆作为单一栽培种植时（轮作种子产量减去 SS 种子产量）分别为 9、20、27、21 和 31 %，分别为 CC/S-CC/S、S-W/S-M、CC/S-W/S-M、M-W/S 和 CC/M-W/S。总之，大豆与谷物轮作的平均增加约 346 公斤公顷-1，并且在大豆单一栽培中包括覆盖作物并没有在与谷物轮作时观察到的水平上增加种子产量。最后，在估计大豆的种子产量差距时，重要的是要确定大豆的轮作位置，因为当大豆进行更密集的轮作时，预计会出现更高的最大产量。