Current Science

B. S. Das ¹, M. C. Sarathjith ¹, P. Santra ², R. N. Sahoo ³, R. Srivastava ⁴, A. Routray ¹, S. S. Ray ⁵

Affiliations
1 Indian Institute of Technology Kharagpur, Kharagpur 721 302, IN
2 Central Arid Zone Research Institute, Jodhpur 342 003, IN
3 Indian Agricultural Research Institute, Pusa, New Delhi 110 012, IN
4 National Bureau of Soil Survey and Land Use Planning, Nagpur 440 033, IN
5 Mahalanobis National Crop Forecast Centre, Pusa Campus, New Delhi 110 012, IN

Abstract

Rapid and reliable assessment of soil characteristics is an important step in agricultural and natural resource management. Over the last few decades, diffuse reflectance spectroscopy (DRS) has emerged as a new tool to obtain both qualitative and quantitative information on soil in a non-invasive manner. The DRS approach is attractive because both the proximal and remote mode of measurements may be adopted to estimate multiple attributes of soil such as physical and chemical soil properties and nutrient contents from a single reflectance spectrum. Hyperspectral imaging cameras onboard remote sensing platforms are already providing hundreds of narrow, contiguous bands of reflectance values and the technology is becoming popular as the hyperspectral remote sensing (HRS) approach. The main objective of this review is to summarize the preparedness and opportunities for using the HRS approach for soil assessment in India. Detailed literature review suggests that the HRS approach requires large spectral databases and robust spectral algorithms in addition to the capability to interpret HRS images. Over the last decade, few efforts have been made to create spectral libraries for Indian soils. However, most of these libraries are very small, precluding the development of robust spectral algorithms. Specifically, the availability of HRS data and robust retrieval algorithms for soil properties from HRS data through unmixing procedures require special attention. With several global initiatives to make HRS data available, coordinated efforts are needed in India to build comprehensive spectral libraries, algorithms and create trained human resources to take full advantage of this emerging technology. Specifically, a dedicated spaceborne mission will provide quality hyperspectral data for the effective application of HRS for soil assessment in India.

Keywords

Hyperspectral Remote Sensing, Reflectance Spectroscopy, Soil Assessment, Spectral Databases and Algorithms.

Full Text

   

M. C. Sarathjith ¹, B. S. Das ², S. P. Wani ³, K. L. Sahrawat ³, A. Gupta ²

Affiliations
1 International Crops Research Institute for the Semi-Arid Tropics, Bamako, BP-320, ML
2 Indian Institute of Technology Kharagpur, Kharagpur 721 302, IN
3 International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Hyderabad 502 324, IN

Abstract

Diffuse reflectance spectroscopy (DRS) operating in wavelength range of 350-2500 nm is emerging as a rapid and non-invasive approach for estimating soil nutrient content. The success of the DRS approach relies on the ability of the data mining algorithms to extract appropriate spectral features while accounting for non-linearity and complexity of the reflectance spectra. There is no comparative assessment of spectral algorithms for estimating nutrient content of Indian soils. We compare the performance of partialleast- squares regression (PLSR), support vector regression (SVR), discrete wavelet transformation (DWT) and their combinations (DWT-PLSR and DWT-SVR) to estimate soil nutrient content. The DRS models were generated for extractable phosphorus (P), potassium (K), sulphur (S), boron (B), zinc (Zn), iron (Fe) and aluminium (Al) content in Vertisols and Alfisols and were compared using residual prediction deviation (RPD) of validation dataset. The best DRS models yielded accurate predictions for P (RPD = 2.27), Fe (RPD = 2.91) in Vertisols and Fe (RPD = 2.43) in Alfisols, while B (RPD = 1.63), Zn (RPD = 1.49) in Vertisols and K (RPD = 1.89), Zn (RPD = 1.41) in Alfisols were predicted with moderate accuracy. The DWT-SVR outperformed all other approaches in case of P, K and Fe in Vertisols and P, K and Zn in Alfisols; whereas, the PLSR approach was better for B, Zn and Al in Vertisols and B, Fe and Al in Alfisols. The DWT-SVR approach yielded parsimonious DRS models with similar or better prediction accuracy than PLSR approach. Hence, the DWT-SVR may be considered as a suitable data mining approach for estimating soil nutrients in Alfisols and Vertisols of India.

Keywords

Diffuse Reflectance Spectroscopy, Discrete Wavelet Transformation, Partial-Least-Squares Regression, Soil Nutrient Contents, Support Vector Regression.

Full Text