Visible and Near Infrared Spectroscopy Combined with Recursive Variable Selection to Quantitatively Determine Soil Total Nitrogen and Organic Matter
JIA Sheng-yao1, 2, TANG Xu3, YANG Xiang-long1, 2, LI Guang4, ZHANG Jian-ming4*
1. School of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China 2. Key Laboratory of Equipment and Informatization in Environment Controlled Agriculture, Ministry of Agriculture, Hangzhou 310058, China 3. Institute of Environment, Resource, Soil and Fertilizer, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China 4. State Key Laboratory of Industrial Control Technology, Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou 310027, China
Abstract：In the present work, recursive variable selection methods (updating both the model coefficients and effective variables during the prediction process) were applied to maintain the predictive abilities of calibration models. This work compared the performances of partial least squares (PLS), recursive PLS (RPLS) and three recursive variable selection methods, namely variable importance in the projection combined with RPLS (VIP-RPLS), VIP-PLS, and uninformative variable elimination combined with PLS (UVE-PLS) for the measurement of soil total nitrogen (TN) and organic matter (OM) using Vis-NIR spectroscopy. The dataset consisted of 195 soil samples collected from eight towns in Wencheng County, Zhejiang Province, China. The entire data set was split randomly into calibration set and prediction set. The calibration set was composed of 120 samples, while the prediction set included 75 samples. The best prediction results were obtained by the VIP-RPLS model. The coefficient of determination (R2) and residual prediction deviation (RPD) were respectively 0.85, 0.86 and 2.6%, 2.7% for soil TN and OM. The results indicate that VIP-RPLS is able to capture the effective information from the latest modeling sample by recursively updating the effective variables. The proposed method VIP-RPLS has the advantages of better performance for Vis-NIR prediction of soil N and OM compared with other methods in this work.
Key words：Visible and near-infrared spectroscopy;Soil total nitrogen;Organic matter;Recursive partial least squares;Recursive variable selection
[1] Chang C, Laird D A. Soil Science, 2002, 167: 110. [2] Vohland M, Besold J, Hill J, et al. Geoderma, 2011, 166: 198. [3] Xie H, Yang X, Drury C F, et al. Canadian Journal of Soil Science, 2011, 91: 53. [4] Ngaard L, Saudland A, Wagner J, et al. Applied Spectroscopy, 2000, 54: 413. [5] Jouan-Rimbaud D, Massart D L, Leardi R, et al. Analytical Chemistry, 1995, 67: 4295. [6] Di Nezio M S, Pistonesi M F, Fragoso W D, et al. Microchemical Journal, 2007, 85: 194. [7] Centner V, Massart D L, de Noord O E, et al. Analytical Chemistry, 1996, 68: 3851. [8] Chong I, Jun C. Chemometrics and Intelligent Laboratory Systems, 2005, 78: 103. [9] Mouazen A M, Maleki M R, De Baerdemaeker J, et al. Soil and Tillage Research, 2007, 93: 13. [10] Wu Y, Jin Y, Li Y, et al. Vibrational Spectroscopy, 2012, 58: 109. [11] Du W, Chen Z, Zhong L, et al. Analytica Chimica Acta, 2011, 690: 64. [12] Dayal B, MacGregor J. Journal of Process Control, 1997, 7: 169. [13] HAN Xiao-ping, WANG Pei-hua, CUI Chuan-jin, et al(韩小平, 王沛华, 催传金, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2010, 26(6): 47. [14] CHU Xiao-li, YUAN Hong-fu, LU Wan-zhen(褚小立, 袁洪福, 陆婉珍). Progress in Chemistry(化学进展), 2004, 16(4): 528. [15] Mu S, Zeng Y, Liu R, et al. Journal of Process Control, 2006, 16: 557. [16] Chen M, Khare S, Zhang H, et al. Industrial and Engineering Chemistry Research, 2013, 52: 7886. [17] Yang H, Kuang B, Mouazen A M. European Journal of Soil Science, 2012, 63: 410. [18] Mouazen A M, De Baerdemaeker J, Ramon H. Journal of Near Infrared Spectroscopy, 2006, 14: 189 [19] Viscarra Rossel R A, Behrens T. Geoderma, 2010, 158: 46.
