In Situ Measurement of Some Soil Properties in Paddy Soil Using Visible and Near-Infrared Spectroscopy

Abstract

In situ measurements with visible and near-infrared spectroscopy (vis-NIR) provide an efficient way for acquiring soil information of paddy soils in the short time gap between the harvest and following rotation. The aim of this study was to evaluate its feasibility to predict a series of soil properties including organic matter (OM), organic carbon (OC), total nitrogen (TN), available nitrogen (AN), available phosphorus (AP), available potassium (AK) and pH of paddy soils in Zhejiang province, China. Firstly, the linear partial least squares regression (PLSR) was performed on the in situ spectra and the predictions were compared to those with laboratory-based recorded spectra. Then, the non-linear least-square support vector machine (LS-SVM) algorithm was carried out aiming to extract more useful information from the in situ spectra and improve predictions. Results show that in terms of OC, OM, TN, AN and pH, (i) the predictions were worse using in situ spectra compared to laboratory-based spectra with PLSR algorithm (ii) the prediction accuracy using LS-SVM (R2>0.75, RPD>1.90) was obviously improved with in situ vis-NIR spectra compared to PLSR algorithm, and comparable or even better than results generated using laboratory-based spectra with PLSR; (iii) in terms of AP and AK, poor predictions were obtained with in situ spectra (R2<0.5, RPD<1.50) either using PLSR or LS-SVM. The results highlight the use of LS-SVM for in situ vis-NIR spectroscopic estimation of soil properties of paddy soils.

In situ near-infrared spectroscopy for soil organic matter prediction in paddy soil, Pasak watershed, Thailand

“…Similar results to those studied with SOM and TN were reported in other studies (Table 3). This study had similar results to other studies that had moderately high accuracy of the predicted results especially in paddy field area in a case study in Malaysia and China (Gholizadech et al 2013, Wenjun et al 2014). In addition, other land use types such as the Mediterranean regime with various land uses in Spain and a wheat crop land use in Canada both had high accuracy of SOM prediction, while the predicted TN result in wheat crop area had low accuracy (Martin et al 2002, Zornoza et al 2008).…”

Application of visible and near infrared reflectance spectroscopy to predict total nitrogen in soil

“…As can be seen in Table 3, 2 R regression value was found as 0,793 and residual prediction 2 deviation value as 2.0802. R and the RPD values calculated in this study for the prediction of total nitrogen were lower than that of previous studies (Xie et al, 2011;Wenjun et al, 2014). Many researchers have reportedsuccessful application of thevis-NIR spectroscopy for the prediction of various soil characteristics.…”

“…In another study, prediction accuracy of total nitrogen with laboratory-based soil spectra using 2 the PLSR method was R of 0.87 and RPD of 2.81 (Wenjun et al, 2014). The successful prediction of soil total nitrogen is mainly due to the nitrogen's direct spectral responses to the overtones and combinations of N-H, C-H+C-H and C-H+C-C in the vis-NIR spectra (Stenberg et al, 2010;Kuang et al, 2012).…”

Possibility of determining soil pH using visible and near-infrared (Vis-NIR) spectrophotometry

“…In their study, they found soil pH measurements taken in the laboratory to be at 2 'Excellent' level of prediction (R =0.85, RMSEP=0.18 and RPD=2.52), while measurements made in the field were found to 2 be 'very good' (R =0.81, RMSEP= 0.20 and RPD=2.14).As a result, it was concluded that visible and infrared spectrophotometry could be used successfully for the estimation of soil pH. Wenjun et al (2014) used vis-NIR spectrophotometry in field and laboratory in order to determine certain soil properties in 2 plowed land. The R value for soil pH was found as 0.82, while the RPD value was 2.42, and thus by looking at these values, they were able to predict the soil pH with approximate quantitative accuracy.…”

“…The RPD evaluation classification for estimating the soil pH performance given in Table 2 suggests that the values of the model succeeded with a moderate level of correlation. According to the reports of Wenjun et al (2014), Williams (2003) and Saeys 2 et al (2005), when the R value was between 0.66 and 0.81 with the RPD value between 2 and 2.5, a 'good' estimation of the examined variable was possible.In a study using visible and nearinfrared spectrophotometry for the purpose of estimating the soil properties of Azerbaijan Mugan soil, the soil pH was found to have 2 a low predictive value according to the obtained R value (Feyziye et al, 2016). Tekin et al (2013) sought to determine the potential of soil pH measurement for deriving the variable rate lime recommendations in the laboratory and on the ground using visible and near infrared spectrophotometry.…”