LAUSR

Abstract Natural rubber ( Hevea brasiliensis ) is a high-valued industrial crop in tropical areas. Reasonable fertilization can improve the quality and production of the natural rubber. Hence, there is a great interest in developing an accurate and robust model to detecting the nitrogen content of rubber tree to improve the production. In this study, we propose a new method for detecting the nitrogen content using fractional calculus augmented NIR spectra. Fractional calculus is applied as a tool to extract additional information from the original spectra. Three different mathematical transforms ( R , 1/ R , log ( R )) are utilized before the fractional order derivative analysis. After derivative analysis with different orders (0–2), the PLS regression method is utilized to develop the estimation models with the wavelengths selected by significant test of correlation coefficient at 0.01 level. The optimal prediction result R P  = 0.9245 is achieved by reciprocal spectrum with the order 0.6. Results show that fractional calculus with NIR can be utilized to build a more accurate nitrogen estimation model for rubber tree. The result presented here provides us a fast and non-destructive strategy to evaluate the nitrogen condition of rubber tree and provides a direct way to support the field management.