There is a growing need for sustainable nutrient management in order to face the many and diverse challenges. These include the excessive use of fertilizer in many first world countries, the depletion of nutrients from soils in third world countries, mainly sub-Saharan Africa, the increasing demand for agricultural products, and the diminishing, non-renewable phosphorus reserves. One step in the process of optimizing nutrient management is predicting the nutrient status of a soil accurately. As a basis for this optimization this thesis describes a methodology to improve the choice of soil tests and interpretation of the corresponding results for phosphorus (P) and potassium (K).
The first step to derive this methodology for P, was to understand the dominant soil processes involved in the translocation of P from soil particles to the plant and the corresponding reaction rates. Experiments were performed wherein the continuous removal of P by a crop was mimicked using an artificial sink. In the experimental setup as used in our study, P uptake by the sink was found to be slower than the release rate by the soil. The desorption rate of the soil was high enough to buffer the P concentration in solution (CP) to a degree that kept this concentration in equilibrium with the soil during the experiment. As a result of this equilibrium, the soil P supply could be described by a soil specific sorption isotherm, relating CP and the reversibly adsorbed P (QP) replenishing CP.
In the second step the gained insight in the dynamic processes is used to derive a combination of standard soil tests that can be used to accurately predict the soil P supply potential. It was found that a minimum of two parameters is needed; a measure for the reversibly adsorbed P (QP, e.g. P-Olsen, PAL) and a measure for CP (e.g. P-CaCl2). The measure for CP is an indication of the rate with which P can be removed from the soil. The ratio QP over CP is an indication of the capacity of the soil to buffer CP. To increase the accuracy of this prediction a measure for the capacity of the soil to adsorb P, i.e. reactive surface area (e.g. Feox and Alox) of the soil must also be taken into account. In addition, to be able to extend the prediction to an amount of P that exceeds the amount of readily desorbable P, a measure for the total amount of P must be included (e.g. Pox).
In the third step it is verified that the same processes that determine the soil P supply potential to an artificial sink are applicable for a growing crop. In addition, the combination of soil tests that is needed to describe this supply can also be used to predict the supply of P from a soil to a growing crop. Results from field trials in the Netherlands show that predicting P uptake by grass is much more accurate when based on the methodology derived in this study compared to the current approach and can be used as a basis for optimizing P fertilizer strategies.
In order to gain a more comprehensive understanding and prediction of the dynamic soil K supply potential a pot experiment was conducted in which the changes in different soil K fractions during uptake by growing grass were studied. To predict the availability of soil K in the short term, a measure for readily exchangeable K (Kexch) appears to be sufficient for all soils. For longer timescales a distinction must be made between soils that can and cannot buffer this Kexch. For the soils in this study a distinction based on soil texture is sufficient. In sandy soils Kexch was not buffered. In the silt and clay soils Kexch was buffered. The absolute contribution of this buffering will depend on the type of crop but an estimation may be based on the ratio of a measure for the amount of K buffering Kexch over a measure for Kexch.
The research described in this thesis has resulted in a well grounded and generally applicable methodology for predicting the potential of soils to supply the nutrients P and K that can be implemented in routine soil laboratories. It thus offers a basis for optimizing nutrient management strategies.
||Het overdadig gebruik van (kunst)mest heeft geleid, en leidt nog steeds, tot milieuproblemen voor oppervlakte- en grondwater. Gevolg hiervan is een groeiende behoefte, alsmede een steeds stringenter mestbeleid, om bemestingsadviezen te optimaliseren waarbij zowel de gewasopbrengst als milieudoelstellingen worden verenigd. Om deze optimalisatie te bereiken is het belangrijk de gekozen meetmethodes en de interpretatie van de resultaten die dat oplevert, beter te onderbouwen. Deze dissertatie presenteert resultaten van onderzoek dat zich richt op de vergroting van het inzicht in de chemische processen die de beschikbaarheid van fosfaat (P) en kalium (K) in de bodem bepalen. Deze inzichten zijn vervolgens gebruikt om een combinatie van standaard bodem meetmethodes af te leiden die in routinelaboratoria gebruikt kunnen worden.