Little is known about the impact of long-term intensive irrigated rice cropping on secondary macro and micronutrients in soils and crops. Therefore, the present study analysed the effect of intensive irrigated rice cropping on nutrient concentrations in soils and rice (grain and straw at harvest) in two long-term experiments that were established in 1991 for intensive rice-based irrigated systems in the Senegal River valley at Ndiaye and Fanaye, Senegal. The experiments included six different fertilizer treatments and rice was grown for two seasons per year. The samples were collected in the 2016/17 dry season (26 years or 52nd continuous rice cropping season after the establishment). Average grain yields from 1991 to 2016/17 for the different fertilizer treatments indicate that yield was limited by N and P at both sites. At Ndiaye, 120 kg N ha⁻¹ seemed sufficient to reach highest achievable yields, whereas 180 kg N ha⁻¹ was necessary at Fanaye. Soil organic carbon and total soil N did increase considerably over the full length of the experiment at both sites and in all treatments even though all crop residues were removed every season. The estimated increase in total topsoil N (across all fertilizer treatments) per season was 10.8 and 11.4 kg N ha⁻¹ at Ndiaye and Fanaye, respectively. Fertilizer treatments differed in their N, P and K rate but significant effects on total soil concentrations were only detected for P. Using the Mehlich3 soil test, we analysed treatment effects on available P, K, Ca, Mg, S, Mn, Zn and Fe. Critical Mehlich3 values from the literature indicated likely P limitation at both sites and in all but one treatment (high PK treatment), a starting but minor K limitation at Fanaye, a strong S limitation at Fanaye, and a minor Zn limitation at Fanaye and Ndiaye. These indications were mostly confirmed by the grain and straw analysis but with the exception of the P limitation, the actual effect of these possible nutrient limitations on production would need to be tested with targeted field experiments. We conclude that continuous rice cultivation in the Sahel for 26 years can be sustained and even increases soil organic carbon and total soil nitrogen. However, in addition to the common N and P deficiencies, likely deficiencies of K, S and Zn are appearing and may begin to limit rice yields in intensive systems in Africa. The Mehlich3 test seems an efficient tool to identify such deficiencies in irrigated rice soils.

关于长期集约化灌溉水稻种植对土壤和作物中次生宏量和微量营养素的影响知​​之甚少。因此，本研究在 1991 年为塞内加尔河流域的以水稻为基础的集约灌溉系统建立的两个长期试验中分析了集约灌溉水稻种植对土壤和水稻（收获时的谷物和秸秆）中养分浓度的影响。在 Ndiaye 和 Fanaye，塞内加尔。实验包括六种不同的肥料处理，水稻每年种植两个季节。样品采集于 2016/17 旱季（建立后第 26 年或第 52 个水稻连作季）。1991 年至 2016/17 年不同施肥处理的平均粮食产量表明，两个地点的产量都受到 N 和 P 的限制。在 Ndiaye，120 kg N ha^-1似乎足以达到可达到的最高产量，而法纳耶需要180 kg N ha ^-1。尽管每个季节都去除了所有作物残留物，但在两个地点和所有处理的整个试验期间，土壤有机碳和土壤总氮确实显着增加。每个季节表土总氮（所有肥料处理）的估计增加量分别为 10.8 和 11.4 kg N ha ^-1分别在 Ndiaye 和 Fanaye。肥料处理的 N、P 和 K 比率不同，但仅检测到 P 对土壤总浓度的显着影响。使用 Mehlich3 土壤测试，我们分析了处理对有效 P、K、Ca、Mg、S、Mn、Zn 和铁。文献中的关键 Mehlich3 值表明，在两个地点和除一种处理（高 PK 处理）之外的所有处理中都可能存在 P 限制，Fanaye 的 K 限制开始但很小，Fanaye 的 S 限制很大，Fanaye 和 Ndiaye 的 Zn 限制很小. 这些迹象大多通过谷物和秸秆分析得到证实，但除磷限制外，这些可能的营养限制对生产的实际影响需要通过有针对性的田间试验进行测试。我们得出的结论是，萨赫勒地区连续 26 年的水稻种植可以持续，甚至可以增加土壤有机碳和土壤总氮。然而，除了常见的 N 和 P 缺乏之外，K、S 和 Zn 的可能缺乏正在出现，并可能开始限制非洲集约化系统中的水稻产量。Mehlich3 测试似乎是识别灌溉水稻土壤中此类缺陷的有效工具。