Fractional abundances study of macronutrients in soil using hyperspectral remote sensing

Abstract

In agriculture, soil fertility is maintained by using the compost, which contains Nitrogen (N), Phosphorus (P) and Potassium (K). Thus, it is required to acquire information about fertility status of soil and to apply the essential amount of composts. The laboratory-based chemical analysis methods for soil macronutrients test can be laborious, time-consuming, cost-intensive and destructive in nature. To overcome these issues, the hyperspectral remote sensing is employed for identification and determination of macronutrients of soil. The objective of this study is spectral unmixing of compositions of soil and NPK compost by using Derivative Analysis for Spectral Unmixing (DASU) approach. The proposed methodology has been tested for soil samples collected from an area located around Roorkee, UK, India. The applied methodology studies the spectral reflectance by using spectroradiometer data. The spectral regions 989.3 nm for pure NPK compost and 2195.1 nm for pure soils have been found optimal spectral absorption features. Accuracy assessment has been carried out on the basis of linear regression model between the true and estimated abundances. The coefficient of determination (R²) values for compositions of silt clay soil and NPK compost has been found at 989.3 nm spectral region as 0.892, 0.897 for compositions of loamy soil and NPK compost and 0.906 for sandy soil and NPK compost. Similarly, R² values obtained at 2195.1 nm spectral region for silt clay soil and NPK compost is 0.932, 0.926 for compositions of loamy soil and NPK compost and 0.933 for sandy soil and NPK compost. The output of this study provides the fractional abundances of compositions of soil and NPK compost. Further, the results have been validated in laboratory by using chemical analysis methods. Thus, it may be concluded that hyperspectral remote sensing may be used in situ to estimate soil fertility status of farm soil.

Keywords:

• Hyperspectral imaging
• spectral derivative
• spectral un-mixing
• soil macronutrients
• precision farming

Acknowledgement

The authors would like to acknowledge Dr. Raja Chowdhury, Assistant Professor, Environmental Engineering Group, Civil Engineering Department, Indian Institute of Technology Roorkee, for his guidance in laboratory-based chemical analysis. The authors would like to thanks anonymous reviewers for their careful reading of our manuscript and their many insightful comments and suggestions for improving the study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Ministry of Human Resource Development, Government of India.