1 / 1

Characterizing Soil Texture Using a Portable X-Ray Fluorescence Spectrometry

Characterizing Soil Texture Using a Portable X-Ray Fluorescence Spectrometry Yuanda Zhu, David C. Weindorf, Wentai Zhang School of Plant, Environmental and Soil Sciences, LSU- AgCenter , Baton Rouge, LA 70803. Introduction. Introduction

elsa
Download Presentation

Characterizing Soil Texture Using a Portable X-Ray Fluorescence Spectrometry

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Characterizing Soil Texture Using a Portable X-Ray Fluorescence Spectrometry Yuanda Zhu, David C. Weindorf, Wentai Zhang School of Plant, Environmental and Soil Sciences, LSU-AgCenter, Baton Rouge, LA 70803 Introduction • Introduction • The development of new technologies for use in field soil survey has produced powerful new quantitative tools for assessing soil physicochemical properties in-situ. • One such technology, portable x-ray fluorescence (PXRF) spectrometry, has shown considerable promise in evaluating elemental concentrations in soils for a wide variety of applications. • Less research is available on how PXRF can be applied to quantifying soil physical properties. • The major objective of this study is to build a simple regression model which can predict the percentages of sand, silt, and clay based on PXRF readings of the soil and can therefore be used to provide soil textural information for fieldwork. Figure 4 Validation of the regression models of sand, silt and clay on the validation sub-datasets from the Louisiana (a, b, and c) and New Mexico (d, e, and f), USA samples (RMSE is root mean squared error), respectively Figure 1 study sites and sampling locations: a. Louisiana and b. Capulin Volcano National Monument, New Mexico, USA. Figure 6 Comparison of soil PXRF readings under the conditions of laboratory (air-dry) and field, using Fe and Rb as examples Conclusions • Material and Methods • The regression models for sand and clay contents of both study sites were developed based on the modeling sub-datasets (two-thirds of each original dataset), indicating that clay and sand percentages can be adequately predicted with PXRF elemental readings, with R2 of 0.862, 0.959, 0.892, and 0.780 for Louisiana sand and clay, and Capulin sand and clay, respectively (Table 1). • The close relationship between Fe and clay, and Rb and clay found in both this study (Figure 3) may imply the models developed, especially clay models, can be applied to larger areas. • The independent validation of the validation sub-datasets confirmed that PXRF readings can be used to estimate soil textural parameters – the percentages of sand, silt and clay, as indicated by the high R2 values of 0.854, 0.682, 0.975, 0.891, 0.875, and 0.876, and low RMSE values of 5.53%, 5.92%, 2.68%, 6.26%, 5.43%, and 2.66%, for Louisiana sand, silt, and clay and Capulin sand, silt, and clay, respectively (Figure 4). In particular, the RMSE values of clay are substantially lower than those reported in previous studies using other proximal sensing techniques. • PXRF showed promising potential to be used as a proximal technique for in-situ determination of soil texture on soil pedons (Figure 5). It should be noted that excessive amount of soil moisture might lower the performance of PXRF models that were based on air-dry samples (Figure 6); the existence of soil moisture exerts a dilution effect on PXRF elemental readings. • Material and Methods • This research study evaluated the feasibility of predicting soil clay and sand content from PXRF data on 584 soil samples collected from highly diverse regions of Louisiana and northeastern New Mexico (Capulin Volcano National Monument) (Figures 1&2). • Louisiana: 55 surface samples (0-12 cm), 288 samples from soil profiles (0-40 cm), and 83 samples from soil pedons (0-100 cm). • New Mexico: 140 surface samples (0-12 cm), and 18 samples from soil pedons (0-100 cm) • An Olympus Innov-X Delta Premium PXRF was used to sequentially scan each soil sample under both field and laboratory conditions assessing 15 elements (K, Ca, Ti, Cr, Mn, Fe, Co, Cu, Zn, As, Rb, Sr, Zr, Ba, and Pb), which were then related to soil textural data processed through traditional laboratory methods with multiple linear regression. Materials and Methods Figure 5 Laboratory measured (+) and portable x-ray fluorescence (PXRF) predicted (-◊-) clay contents for five pedons from Louisiana, USA Table 1 Backward step-wise multiple linear regression models for sand and clay percentages of the modeling sub-datasets from both Louisiana and New Mexico (Capulin) samples, USA Figure 2 Soil texture of the Louisiana samples (left) Capulin samples (right). Figure 3 The relationship between a.) Fe and clay, and b.) Rb and clay of soil samples

More Related