Variability of clay mineralogy in two New Zealand steep-land topsoils under pasture

The variability of the clay mineralogy in the topsoil of agricultural fields has rarely been examined, although it is important to the construction of accurate potassium (K) availability models. The variability of clay mineralogy and the relationship to plant available nonexchangeable K (Step K) were examined in two steep-land fields in the southern North Island of New Zealand. The soils at the sites have developed from quartzo-feldspathic, mica-bearing, parent materials.The mineralogy was examined using X-ray diffraction (XRD) and differential thermal analysis (DTA). The clay mineralogy consisted of mixtures of mica, interlayered mica-smectite (MS), vermiculite, interlayered mica-vermiculite (MV), hydroxy interlayered vermiculite (HIV), smectite, quartz, feldspar and kaolin minerals. The C.V. of the content of 2:1 layer silicates ranged from 52% to 257%, indicating large variability within the fields. Correspondence analysis identified an antagonistic relationship, and a large negative correlation, between the proportion of mica and MS compared to the proportion of vermiculite, HIV, MV and chlorite in the clay fraction. In most locations, there was a logical relationship between the mineralogy and the position of the soil in the landscape, as an indicator of the extent of weathering. An exception was the clay mineralogy of topsoils under well-drained and heavily frequented animal campsites, which was characterized by large mica and MS contents, a lack of HIV, and irregularly interstratified expanded 2:1 layer silicates. A reversal of the 2:1 silicate weathering processes was indicated. The large spatial variability of the clay mineralogy, and the evidence of modifications to the mineralogy on the animal campsites, indicated that fixed values based on reference profiles from Soil Orders would inaccurately represent the effect of clay mineralogy in K availability models. Clay mineralogy is expensive to measure; however, Step K values were highly correlated to the 2:1 silicate clay mineralogy. Step K can therefore be sampled within a field to cost-effectively represent clay mineralogy trends in K availability models.