Availability of soil potassium and diagnostic soil tests

Thirty-seven surface (0-0.10 or 0-0.20 m) soils covering a wide range of soil types (16 Vertosols, 6 Ferrosols, 6 Dermosols, 4 Hydrosols, 2 Kandosols, 1 Sodosol, 1 Rudosol, and 1 Chromosol) were exhaustively cropped in 2 glasshouse experiments. The test species were Panicum maximum cv. Green Panic in Experiment A and Avena sativa cv. Barcoo in Experiment B. Successive forage harvests were taken until the plants could no longer grow in most soils because of severe potassium (K) deficiency. Soil samples were taken prior to cropping and after the final harvest in both experiments, and also after the initial harvest in Experiment B. Samples were analysed for solution K, exchangeable K (Exch K), tetraphenyl borate extractable K for extraction periods of 15 min (TBK15) and 60 min (TBK60), and boiling nitric acid extractable K (Nitric K). Inter-correlations between the initial levels of the various soil K parameters indicated that the following pools were in sequential equilibrium: solution K, Exch K, fast release fixed K [estimated as (TBK15-Exch K)], and slow release fixed K [estimated as (TBK60-TBK15)]. Structural K [estimated as (Nitric K-TBK60)] was not correlated with any of the other pools. However, following exhaustive drawdown of soil K by cropping, structural K became correlated with solution K, suggesting dissolution of K minerals when solution K was low. The change in the various K pools following cropping was correlated with K uptake at Harvest 1 ( Experiment B only) and cumulative K uptake ( both experiments). The change in Exch K for 30 soils was linearly related to cumulative K uptake (r = 0.98), although on average, K uptake was 35% higher than the change in Exch K. For the remaining 7 soils, K uptake considerably exceeded the change in Exch K. However, the changes in TBK15 and TBK60 were both highly linearly correlated with K uptake across all soils (r = 0.95 and 0.98, respectively). The slopes of the regression lines were not significantly different from unity, and the y-axis intercepts were very small. These results indicate that the plant is removing K from the TBK pool. Although the change in Exch K did not consistently equate with K uptake across all soils, initial Exch K was highly correlated with K uptake (r = 0.99) if one Vertosol was omitted. Exchangeable K is therefore a satisfactory diagnostic indicator of soil K status for the current crop. However, the change in Exch K following K uptake is soil-dependent, and many soils with large amounts of TBK relative to Exch K were able to buffer changes in Exch K. These soils tended to be Vertosols occurring on floodplains. In contrast, 5 soils (a Dermosol, a Rudosol, a Kandosol, and 2 Hydrosols) with large amounts of TBK did not buffer decreases in Exch K caused by K uptake, indicating that the TBK pool in these soils was unavailable to plants under the conditions of these experiments. It is likely that K fertiliser recommendations will need to take account of whether the soil has TBK reserves, and the availability of these reserves, when deciding rates required to raise exchangeable K status to adequate levels.

Simple software processes and tests improve the reliability and usefulness of a model.

Models are abstractions of reality that have predetermined limits (often not consciously thought through) on what problem domains the models can be used to explore. These limits are determined by the range of observed data used to construct and validate the model. However, it is important to remember that operating the model beyond these limits, one of the reasons for building the model in the first place, potentially brings unwanted behaviour and thus reduces the usefulness of the model. Our experience with the Agricultural Production Systems Simulator (APSIM), a farming systems model, has led us to adapt techniques from the disciplines of modelling and software development to create a model development process. This process is simple, easy to follow, and brings a much higher level of stability to the development effort, which then delivers a much more useful model. A major part of the process relies on having a range of detailed model tests (unit, simulation, sensibility, validation) that exercise a model at various levels (sub-model, model and simulation). To underline the usefulness of testing, we examine several case studies where simulated output can be compared with simple relationships. For example, output is compared with crop water use efficiency relationships gleaned from the literature to check that the model reproduces the expected function. Similarly, another case study attempts to reproduce generalised hydrological relationships found in the literature. This paper then describes a simple model development process (using version control, automated testing and differencing tools), that will enhance the reliability and usefulness of a model.