Potential for spectral indices to remotely sense phosphorus and potassium content of legume-based pasture as a means of assessing soil phosphorus and potassium fertility status

Abstract

Precision nutrient management needs analytical tools that aid collection of site-specific data. Adequate soil phosphorus (P) and potassium (K) fertility is crucial for pasture production in New Zealand. This article explores (a) the relationship between 12 spectral indices from in situ canopy reflectance and pasture growth rate (PGR), and pasture P and K content in pastures, (b) the performance of the model in different seasons and (c) the relationship between sensed pasture P and K content and soil P (Olsen P) and K (exchangeable K) fertility. Hyperspectral data were collected from a small area of each of 30 legume-based pastures that varied in soil P (Olsen P 5–72 mg kg⁻¹) and soil exchangeable K (0.20−1.32 cmol kg⁻¹) in spring 2004 and again in summer 2006. Overall, the photochemical reflectance index (PRI) showed the best coefficients of determination (R ²) for most variables. In an exploratory analysis using all the spectral waveband data, normalized difference spectral indices (NDSIs) using the combination of reflectance at 523 and 583 nm of the pasture canopy gave the best prediction of soil P and exchangeable K status. The prediction of Olsen P from plant P (R ² > 0.89) and soil K from plant K (R ² > 0.73) was achieved through fitted logarithmic functions that linked plant P and K to soil P and K status, respectively. This pilot study has been broadened to examine other methodologies for interpreting the spectral data and extended to other pasture types and soil orders.

Acknowledgements

We thank Roger Parfitt (Landcare Research) for access to the plots and data from the biodiversity study. This work was supported by a Research Fellowship of the Japan Society for the Promotion of Science (JSPS) for Young Scientists, and funding from the FRST contract C02X0405 (Sustainable Land Use Research Initiative) and ARC05031 (Satellites on farms) research programmes.