Portable X-ray fluorescence(pXRF) spectrometers can be used to determine the elemental composition easily, rapidly, and without using chemical reagents, which is very important for tropical regions due to the lack of ...Portable X-ray fluorescence(pXRF) spectrometers can be used to determine the elemental composition easily, rapidly, and without using chemical reagents, which is very important for tropical regions due to the lack of detailed soil characterization data. Moreover,pXRF data can be used to predict the results of more expensive, time-consuming, and conventional laboratory analyses. This study sought to determine the elemental composition of various soil profiles using pXRF. Two operational modes(Trace Mode and General Mode) and two scanning time(30 and 60 s) were assessed to determine their effects on the correlation of pXRF dataset with respect to conventional inductively coupled plasma(ICP)-optical emission spectrometry analysis. This relationship has been reported in previous studies, however, few studies were performed on tropical soils, which are unique. Furthermore, such relationships establish the viability of developing prediction models directly from pXRF data. Linear regression was applied to develop calibration models for the prediction of ICP analysis results and exchangeable and available elemental contents based on pXRF data. High coefficients of determination(R^2) were obtained for Ca(0.87), Cu(0.90), Fe(0.95), Mn(0.85), Cr(0.95), V(0.72), and Ni(0.90), with adequate validation. Statistically significant results were not found for Al, K, Zn, Ti, and Zr. The models predicting the exchangeable Ca based on the total Ca from p XRF reached an R^2 of up to 0.85. Operational modes influenced the pXRF results. Our results illustrate that pXRF holds great promise for tropical soil characterization and the development of prediction models, justifying the need for larger-scale studies in tropical countries worldwide.展开更多
Soils result from the interaction of five independent formation factors.If one factor varies,while the others remain constant,different soils can be produced.Herein,we demonstrated an opposing trend,wherein two soils ...Soils result from the interaction of five independent formation factors.If one factor varies,while the others remain constant,different soils can be produced.Herein,we demonstrated an opposing trend,wherein two soils were similar,despite considerable differences in all factors of soil formation.We sampled two Inceptisols(Oxic Dystrudepts) formed on different parent materials(gneiss vs.mica schist),climate(tropical altimontane vs.warmer,drier plateau),topography(1 650 m,45% slope vs.1 000 m,8% slope),time(rejuvenated vs.old,stable surface),and vegetation(rainforest vs.Cerrado savanna).The two soils had similar chemical properties,whereas the soil on mica schist had finer particle size distribution,lower porosity,and lower saturated hydraulic conductivity.These properties were related to a coarser blocky microstructure compared to the soil on gneiss.Both soils presented active mineral weathering and pronounced pedoplasmation,demonstrated by clay contents>300 g kg^(-1),although only the Dystrudept on gneiss possessed coarse rock fragments.The C horizons of both soils presented fragmented clay coatings suggestive of argilluviation,likely relict,because they were not observed in the B horizons.The similarities in many properties of the two Dystrudepts,despite contrasting factors of soil formation,suggest converging evolution and that soil classification at the subgroup level was efficient in grouping similar formative processes in tropical conditions.Moreover,this work revealed that similar pedogenic processes acting on different factors of soil formation can result in similar soil properties,at least for Inceptisols where further soil development is hindered by topographic limitations.展开更多
Portable X-ray fluorescence(pXRF) spectrometry and magnetic susceptibility(MS) via magnetometer have been increasingly used with terrain variables for digital soil mapping. However, this methodology is still emerging ...Portable X-ray fluorescence(pXRF) spectrometry and magnetic susceptibility(MS) via magnetometer have been increasingly used with terrain variables for digital soil mapping. However, this methodology is still emerging in many countries with tropical soils. The objective of this study was to use proximal soil sensor data associated with terrain variables at varying spatial resolutions to predict soil classes using the Random Forest(RF) algorithm. The study was conducted on a 316-ha area featuring highly variable soil classes and complex soil-landscape relationships in Minas Gerais State, Brazil. The overall accuracy and Kappa index were evaluated using soils that were classified at 118 sites, with 90 being used for modeling and 28 for validation. Digital elevation models(DEMs) were created at 5-, 10-, 20-, and 30-m resolutions using contour lines from two sources. The resulting DEMs were processed to generate 12 terrain variables. Total Fe, Ti, and SiO_(2) contents were obtained using pXRF, with MS determined via a magnetometer. Soil class prediction was performed using the RF algorithm. The quality of the soil maps improved when using only the five most important covariates and combining proximal sensor data with terrain variables at different spatial resolutions. The finest spatial resolution did not always provide the most accurate maps. The high soil complexity in the area prevented highly accurate predictions. The most important variables influencing the soil mapping were MS, Fe, and Ti. Proximal sensor data associated with terrain information were successfully used to map Brazilian soils at variable spatial resolutions.展开更多
基金the Brazilian funding agenciesNational Council for Scientific and Technological Development(CNPq)+1 种基金Coordination of Superior Level Staff Improvement(CAPES)Foundation for Research of the State of Minas Gerais(FAPEMIG)
文摘Portable X-ray fluorescence(pXRF) spectrometers can be used to determine the elemental composition easily, rapidly, and without using chemical reagents, which is very important for tropical regions due to the lack of detailed soil characterization data. Moreover,pXRF data can be used to predict the results of more expensive, time-consuming, and conventional laboratory analyses. This study sought to determine the elemental composition of various soil profiles using pXRF. Two operational modes(Trace Mode and General Mode) and two scanning time(30 and 60 s) were assessed to determine their effects on the correlation of pXRF dataset with respect to conventional inductively coupled plasma(ICP)-optical emission spectrometry analysis. This relationship has been reported in previous studies, however, few studies were performed on tropical soils, which are unique. Furthermore, such relationships establish the viability of developing prediction models directly from pXRF data. Linear regression was applied to develop calibration models for the prediction of ICP analysis results and exchangeable and available elemental contents based on pXRF data. High coefficients of determination(R^2) were obtained for Ca(0.87), Cu(0.90), Fe(0.95), Mn(0.85), Cr(0.95), V(0.72), and Ni(0.90), with adequate validation. Statistically significant results were not found for Al, K, Zn, Ti, and Zr. The models predicting the exchangeable Ca based on the total Ca from p XRF reached an R^2 of up to 0.85. Operational modes influenced the pXRF results. Our results illustrate that pXRF holds great promise for tropical soil characterization and the development of prediction models, justifying the need for larger-scale studies in tropical countries worldwide.
基金the Brazilian funding agencies Capes/Fapemig(No.23038.008715/2012-21)Fapemig(No.CAG PPM 00132/14)Conselho Nacional de Pesquisa(No.303840/2014-5)
文摘Soils result from the interaction of five independent formation factors.If one factor varies,while the others remain constant,different soils can be produced.Herein,we demonstrated an opposing trend,wherein two soils were similar,despite considerable differences in all factors of soil formation.We sampled two Inceptisols(Oxic Dystrudepts) formed on different parent materials(gneiss vs.mica schist),climate(tropical altimontane vs.warmer,drier plateau),topography(1 650 m,45% slope vs.1 000 m,8% slope),time(rejuvenated vs.old,stable surface),and vegetation(rainforest vs.Cerrado savanna).The two soils had similar chemical properties,whereas the soil on mica schist had finer particle size distribution,lower porosity,and lower saturated hydraulic conductivity.These properties were related to a coarser blocky microstructure compared to the soil on gneiss.Both soils presented active mineral weathering and pronounced pedoplasmation,demonstrated by clay contents>300 g kg^(-1),although only the Dystrudept on gneiss possessed coarse rock fragments.The C horizons of both soils presented fragmented clay coatings suggestive of argilluviation,likely relict,because they were not observed in the B horizons.The similarities in many properties of the two Dystrudepts,despite contrasting factors of soil formation,suggest converging evolution and that soil classification at the subgroup level was efficient in grouping similar formative processes in tropical conditions.Moreover,this work revealed that similar pedogenic processes acting on different factors of soil formation can result in similar soil properties,at least for Inceptisols where further soil development is hindered by topographic limitations.
基金BL Allen Endowment in Pedology at Texas Tech University,USAthe Brazilian funding agencies National Council for Scientific and Technological Development (CNPq) (Nos.301930/2019-8 and 306389/2019-7)+1 种基金the Coordination for the Improvement of Higher Education Personnel (CAPES),Brazil (No.590-2014)Research Support Foundation of the State of Minas Gerais (FAPEMIG),Brazil (No.PPM 00305-17) for the financial support provided。
文摘Portable X-ray fluorescence(pXRF) spectrometry and magnetic susceptibility(MS) via magnetometer have been increasingly used with terrain variables for digital soil mapping. However, this methodology is still emerging in many countries with tropical soils. The objective of this study was to use proximal soil sensor data associated with terrain variables at varying spatial resolutions to predict soil classes using the Random Forest(RF) algorithm. The study was conducted on a 316-ha area featuring highly variable soil classes and complex soil-landscape relationships in Minas Gerais State, Brazil. The overall accuracy and Kappa index were evaluated using soils that were classified at 118 sites, with 90 being used for modeling and 28 for validation. Digital elevation models(DEMs) were created at 5-, 10-, 20-, and 30-m resolutions using contour lines from two sources. The resulting DEMs were processed to generate 12 terrain variables. Total Fe, Ti, and SiO_(2) contents were obtained using pXRF, with MS determined via a magnetometer. Soil class prediction was performed using the RF algorithm. The quality of the soil maps improved when using only the five most important covariates and combining proximal sensor data with terrain variables at different spatial resolutions. The finest spatial resolution did not always provide the most accurate maps. The high soil complexity in the area prevented highly accurate predictions. The most important variables influencing the soil mapping were MS, Fe, and Ti. Proximal sensor data associated with terrain information were successfully used to map Brazilian soils at variable spatial resolutions.
