Mineral nutrition of crops along with N, P, and K fertilization decisions in cropping systems have long been based on crop yield responses to individual elements for determining the optimum rate of application of fertilizers for maximum yield. Owing to the great uncertainty for estimating or forecasting these response curve parameters, farmers often apply fertilizers in excess to avoid any risk of yield reduction. Undesirable environmental impacts have resulted from this over-fertilization. Therefore, the old paradigm based on prognosis of soil nutrient availability estimated by soil tests must be deeply questioned and reappraised for a more sustainable and efficient agro-ecosystem management.

For N fertilization management, attempts have been made for reducing the uncertainty of the prognosis-based approach using the crop N balance concept but with a relatively poor success because of the difficulty in forecasting soil N supply and crop N demand in dynamic terms. A new “plant-based diagnosis” approach of crop N nutrition using the concept of “critical N dilution curve” and well sustained by a strong theoretical framework has been proposed. This concept has been extended successfully to a large range of arable crops. This new approach allows the monitoring of crop N status in real time by detecting if the N supply matches or not crop N demand. This in situ crop N diagnosis should help determine when and how much N fertilization is required and could be incorporated within crop N fertilization management procedures to be used by farmers.

More recent research has shown that the same theoretical framework is also relevant for P, K and S crop nutrition. Along the opportunity of developing plant-based diagnosis of those three single nutrients, it provides an overall framework for developing an integrated crop N-P -K-S nutrition diagnosis for a balanced crop fertilization management.

This paper introduces a special issue devoted to the development and implementation of this crop nutrition diagnosis approach. The papers in this special issue highlight the significant progress made over the last thirty years along with providing an insight and perspective of future research needs.

长期以来，农作物的矿物质营养以及氮，磷和钾的施肥决策一直基于作物对单个元素的产量响应，以确定最大产量的最佳肥料用量。由于估计或预测这些响应曲线参数存在很大的不确定性，因此农民经常过量施肥，以避免任何减产的风险。这种过度施肥对环境造成了不良影响。因此，必须对基于土壤试验估算的土壤养分可利用性的预测的旧模式进行深入质疑和重新评估，以实现更可持续，更有效的农业生态系统管理。

对于氮肥管理，已经尝试使用作物氮平衡的概念来减少基于预测的方法的不确定性，但由于以动态方式预测土壤氮供应和作物氮需求的困难，取得了相对较差的成功。提出了一种新的基于作物氮素营养的“基于植物的诊断”方法，该方法以“临界氮稀释曲线”为概念，并通过强大的理论框架得到了很好的维持。这个概念已经成功地扩展到了许多耕作作物上。这种新方法允许通过检测氮素供应是否匹配来实时监测作物氮素状况。这原位 作物氮素诊断应有助于确定何时以及需要多少氮肥，并且可以纳入农民使用的作物氮素管理程序中。

最近的研究表明，相同的理论框架也与磷，钾和硫作物营养有关。随着开发基于植物的这三种单一营养素诊断的机会，它为开发综合作物NP -KS营养诊断提供了一个总体框架，从而实现了平衡的作物施肥管理。

本文介绍了专门针对这种作物营养诊断方法的开发和实施的特刊。本期特刊的论文着重介绍了过去三十年中取得的重大进展，并提供了对未来研究需求的见解和观点。