DIVERSE Project Field Experiment Dataset

A field plot experiment was located on the University of Reading farm at Sonning in Berkshire, UK (51°28'21"N -0°54'13"E) and was established in August 2018. Plots were 2 x 11m in size, arranged in a randomised complete block design with 4 blocks/replicates. Each summer, after the harvest of a cereal crop, cover crops were planted and then mowed in the autumn using a lawn mower and either incorporated into the soil of the same plot (Mulch), or removed from the plot (Remove) and added to a plot where cover crops were not grown (Add). Cover crop residues were incorporated by ploughing prior to planting the next autumn sown cereal crop. The cereal rotation was Winter Wheat (2017/18); Winter Barley (2018/19); Winter Oats (2019/20); and Winter Wheat (2020/21). The cover crop treatments were control (no cover crops), buckwheat (Fagopyrum esculentum; variety Hajnalka), berseem clover (Trifolium alexandrinum; variety TIM), oil radish (Raphanus raphanistrum; variety Barracuda), sunflower (Helianthus annuus; variety M6103), and a 4-species mixture of these four. Nitrogen fertiliser was applied at a rate equivalent to 75% of the AHDB RB209 recommended rate for the crop. All other fertiliser and crop protection was applied in line with standard farm practice.

Timeline:
Cover crops drilled on 14th August 2018
Cover crops terminated on 10th October 2018
Barley drilled on 23rd October 2018
Barley harvested on 12th and 15th July 2019
Cover crops drilled on 12th August 2019
Cover crops terminated on 21st October 2019
Oats drilled on 29th October 2019
Oats harvested on 24th August 2020
Cover crops drilled on 1st September 2020
Cover crops terminated on 2nd November 2020
Wheat drilled on 6th November 2020
Wheat harvested on 13th August 2021

Soil samples were taken as a composite sample of three gauge auger cores per plot to a depth of 0-20 cm each summer after cereal crop harvest and before cover crop planting. Soil samples were sieved moist to 4 mm and analysed for soil moisture and available nitrate and ammonium. The remainder of the soil was air dried and sieved to 2 mm and analysed for organic matter, soil pH, Olsen P, and EDTA extractable Cu, Fe, K, Mg, Mn, Na, Ni, S, and Zn. Dry soil was ball milled to a fine powder and analysed for total C and N. Samples (air dried and 2 mm sieved) taken in 2018 only were analysed for soil texture and total (reverse aqua regia digestable) Ca, Cu, Fe, K, Mg, Mn, Na, Ni, P, S, and Zn.

Grain yield was measured using a plot combine and expressed after moisture correction in dry kg per Ha. Grain was milled and analysed for total N, C, Ca, Cu, Fe, K, Mg, Mn, Mo, Na, Mi, P, S, and Zn.
The statistical design of the experiment allows analysis using the following nested ANOVA model: Cover/Treatment/Mixture/Species. Cover has two levels (Yes and No). Treatment has four levels (Mulch, Add, Remove, and Control). Mixture has two levels (Yes and No), and Species has six levels (Control, Buck, Sun, Clover, Radish, and Mix).

一项田间试验实验位于英国伯克郡桑宁的雷丁大学农场（北纬51°28'21''，东经0°54'13''），该实验于2018年8月建立。试验地块面积为2 x 11米，采用随机完全区组设计，共分为4个区组/重复。每年夏季，在收获谷物作物后，种植覆盖作物，并在秋季使用割草机将其刈割，然后将其（1）混合入同一地块的土壤中（覆盖），或（2）从地块中移除并添加到未种植覆盖作物的地块（添加）。在种植下一个秋季播种的谷物作物之前，通过耕作将覆盖作物残留物混合入土壤。谷物轮作顺序为：冬季小麦（2017/18年）；冬季大麦（2018/19年）；冬季燕麦（2019/20年）；冬季小麦（2020/21年）。覆盖作物处理包括：对照（无覆盖作物）、荞麦（Fagopyrum esculentum，品种Hajnalka）、草木樨（Trifolium alexandrinum，品种TIM）、油菜籽（Raphanus raphanistrum，品种Barracuda）、向日葵（Helianthus annuus，品种M6103）以及这四种作物混合的四种组合。氮肥的施用量相当于AHDB RB209推荐施肥量的75%。所有其他肥料和作物保护措施均按照标准农场实践进行。

时间线：
覆盖作物于2018年8月14日播种
覆盖作物于2018年10月10日终止
大麦于2018年10月23日播种
大麦于2019年7月12日和15日收获
覆盖作物于2019年8月12日播种
覆盖作物于2019年10月21日终止
燕麦于2019年10月29日播种
燕麦于2020年8月24日收获
覆盖作物于2020年9月1日播种
覆盖作物于2020年11月2日终止
小麦于2020年11月6日播种
小麦于2021年8月13日收获

土壤样本取自每个地块三个量规钻头孔的混合样，深度为每块0-20厘米。每年在谷物作物收获后和覆盖作物种植前采集土壤样本，将其筛至4毫米并分析土壤水分、可利用硝酸盐和铵。剩余的土壤样本经风干后筛至2毫米，并分析有机物、土壤pH、Olsen P以及EDTA可提取的Cu、Fe、K、Mg、Mn、Na、Ni、S和Zn。风干的土壤样本经球磨成细粉，并分析总碳和总氮。仅2018年采集的空气干燥并筛至2毫米的样本分析了土壤质地和总（反硝酸消化）Ca、Cu、Fe、K、Mg、Mn、Na、Ni、P、S和Zn。

谷物产量使用地块联合收割机测量，经水分校正后以干公斤每公顷表示。谷物经磨碎后分析总氮、碳、钙、铜、铁、钾、镁、锰、钼、钠、钴、磷、硫和锌。
实验的统计设计允许使用以下嵌套方差分析模型进行分析：覆盖/处理/混合/物种。覆盖有两个水平（是/否）。处理有四个水平（覆盖、添加、移除和对照）。混合有两个水平（是/否），物种有六个水平（对照、荞麦、向日葵、草木樨、萝卜和混合）。