Plant available potassium in New Zealand steep-land pasture soils

Abstract

In order to examine the effect of the long-term animal transfer of nutrients, soil K fertility patterns were investigated in two established steep-land pastures in the southern North Island of New Zealand. Spatially randomized and transect style surveys were made of NH₄-exchangeable K and plant available nonexchangeable K (Step K) values. Wide-ranging and highly skewed distributions of values were found. The large short-range spatial variability could only form nugget variograms, so that mapping was not possible, however correspondence analysis was useful to examine the distribution of values obtained at different landscape positions. The range of NH₄-exchangeable K values increased markedly on ridge-tops in association with well-drained animal campsites. Plant available Olsen P values had a similar pattern. Step K values were affected by a mixture of soil parent material and soil moisture effects, including an interaction between soil moisture and animal camping patterns. Although the animal transfer effects were extensive on the campsite areas, no accompanying reduction of the K fertility on areas of the steeper slopes was found. Analysis of the total K and P content of the topsoils indicated a relatively low rate of transfer of K from the slopes to the ridge-tops. Consequently, the development of significant K deficiencies in the soil on the slopes is not expected. Current soil sampling methods in steep-land usually make estimates from soil samples that are collected over large areas and then combined (composited/bulked). In this study, the wide ranges of values and skewed distributions indicated that more meaningful estimates of K status could be made if a population distribution of K fertility values was described using individual samples.

Introduction

The transfer of nutrients by stock from the steep slopes to the flatter areas of fields is a well-established process in grazed hilly pastures. The loss effects can accumulate to cause nutrient deficiencies on the slopes in affected pastures (Gillingham and During (1973). Concerns have arisen that potassium (K) deficiencies may be developing as a result of the transfer process in some areas of steep-land pasture in the southern North Island of New Zealand. The soil in these pastures has been previously classified as containing adequate reserves of K and K has not normally been applied in the fertilizer. Consequently, in the thirty or more years that these soils have been farmed, K deficiencies may have developed on the slopes.

The lowland steep-lands of New Zealand, popularly called the “hill country”, are steep non-arable hills below 1000 m, which cover more than 40% of the country's land surface. Steep-land terrain is characterized by rough topography and rapid changes from water shedding (convex) to water collecting (concave) land formations. Slopes are defined as moderately steep when the ground slope is between 21° and 25°, steep between 26° and 35°, and very steep when the ground slope is > 35° (NZLRI, 1979). The term steep-land also implies that soil catenas or complexes are present, so that several distinct soil profile classes are distributed in either a regular or an undetermined pattern. The original forest typically has been removed some time in the last 60–130 years. A pasture has developed that is a mixture of low producing grasses and weeds on the steep areas, and a ryegrass (Lolium perrenne) and white clover (Trifolium repens) mixture in the more fertile flatter areas. The grazing stock are often a mixture of cattle and sheep, which modify the terrain by the formation of tracks on the small terracettes that follow the contour (Tonkin et al., 1982, Blaschke et al., 1992).

Transfer losses are created when the grazing stock eat the pasture on the slopes and then congregate on more favored parts of a field. As a consequence, over half the total dung and urine excreted may be concentrated on only 15–30% of the area of a field (Haynes and Williams, 1993). The favored areas, which are marked by accumulations of dung, are termed campsites. These areas are characterized by a general increase in fertility and in pasture yields, while, in turn, distinct areas on the slopes are reduced in fertility by the transfer losses. Studies of the transfer effect, which, in New Zealand, have mainly dealt with soil P, have shown wide variations in the actual extent of nutrient transfer from the slopes to the campsites (Gillingham and During, 1973, Rowarth et al., 1992, Saggar et al., 1988).

The effect of K transfer by grazing stock has not been examined in detail in New Zealand steep-land pastures with soil formed in sedimentary parent rock although some responses to K fertilizer have been recorded on these and similar soils (During, 1984, Ledgard et al., 1997). In the southern North Island of New Zealand, the steep-lands are generally formed on deeply dissected quartzo-feldspathic sedimentary parent rock that contains mica (Wells and Furkert, 1973). The soils have been classified as containing K reserves that are adequate to sustain long term pasture growth, so that K fertilizers are not normally applied (Metson, 1980, During, 1984). However, concerns have arisen in New Zealand as to the possible development of K deficiencies in some areas of these steep-land fields, caused by small but constant animal transfer losses of K over time. The effect has been previously established in hilly pastures that have soils formed in volcanic parent materials that contain little K (Gillingham and During, 1973). In this study, K fertility patterns were examined in typical long-established southern North Island steep-land pastures. A survey style approach was used to characterize the cumulative effects of the transfer of K that has occurred during many years of rotational grazing management, where the pastures are grazed at semi-regular intervals as sufficient feed becomes available. The pastures on the farms examined have been established and regularly fertilized with P and S for at least 30 years, but K fertilizer has not been historically applied.

In this study, both plant available exchangeable and nonexchangeable K were examined. Previous studies of K transfer have examined the effect in terms of exchangeable K however, the effect on the nonexchangeable pool has not been examined.