Abstract
Diffuse reflectance spectroscopy (DRS) is emerging as a rapid and cost-effective alternative to routine laboratory analysis for many soil properties. However, it has primarily been applied in project-specific contexts. Here, we provide an assessment of DRS spectroscopy at the scale of the continental United States by utilizing the large (n > 50,000) USDA National Soil Survey Center mid-infrared spectral library and associated soil characterization database. We tested and optimized several advanced statistical approaches for providing routine predictions of numerous soil properties relevant to studying carbon cycling. On independent validation sets, the machine learning algorithms Cubist and memory-based learner (MBL) both outperformed random forest (RF) and partial least squares regressions (PLSR) and produced excellent overall models with a mean R2 of 0.92 (mean ratio of performance to deviation = 6.5) across all 10 soil properties. We found that the use of root-mean-square error (RMSE) was misleading for understanding the actual uncertainty about any particular prediction; therefore, we developed routines to assess the prediction uncertainty for all models except Cubist. The MBL models produced much more precise predictions compared with global PLSR and RF. Finally, we present several techniques that can be used to flag predictions of new samples that may not be reliable because their spectra fall outside of the calibration set.
Highlights
Soil is an essential part of the natural environment that influences the distribution of plants, animals and landforms and plays key roles in providing ecosystem services necessary for mankind, including climate regulation, soil fertility and fiber and food production [1,2]
We focused on 10 soil physical and chemical properties, including organic carbon (OC, %), which was measured as total carbon by elemental analysis minus any inorganic carbon measured manometrically; calcium carbonate (CO3, %) measured by manometer; cation exchange capacity (CEC, cmolc kg−1 ) and exchangeable calcium (Ca, cmolc kg−1 ) measured by displacement with ammonium oxalate, buffered at pH 7; clay (%) by sedimentation, pH in 1:1 water suspension; bulk density (BD, g/cm−3 ); dithionite citrate extractable aluminum (Al, %); acid oxalate extractable iron (Fe, %); and organic carbon density (OCD, kg m−3 ) calculated as the product of OC × best (OC) and worst (BD) ×
On the basis of this analysis, we found that partial least squares regression (PLSR) models built with square-root-transformed analytical data were, on average, superior compared with box-cox-transformed and untransformed data, when PLSR calibration models were applied to independent validation sets (Table S1; Figure S2)
Summary
Soil is an essential part of the natural environment that influences the distribution of plants, animals and landforms and plays key roles in providing ecosystem services necessary for mankind, including climate regulation, soil fertility and fiber and food production [1,2]. Anthropogenic activities have greatly altered the composition and functioning of soils [3,4]. Quantifying the impact of anthropogenic activities on soil carbon sequestration and loss requires at least sporadic monitoring of soil physical, chemical and biological properties that are most relevant to controlling soil carbon cycling rates. Current technologies for monitoring and characterizing most soil properties are expensive and often time-consuming. The total cost of standard soil characterization procedures at the US National Soil Survey Center is about $2500 per pedon with processing times of. There is an increasing need to develop rapid and cost-effective techniques to characterize soil resources
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
