Geophysical sensors combined with machine learning
algorithms were used to understand the pedosphere system and landscape
processes and to model soil attributes. In this research, we used parent
...material, terrain attributes, and data from geophysical sensors in different
combinations to test and compare different and novel machine learning
algorithms to model soil attributes. We also analyzed the importance of
pedoenvironmental variables in predictive models. For that, we collected
soil physicochemical and geophysical data (gamma-ray emission from uranium,
thorium, and potassium; magnetic susceptibility and apparent electric
conductivity) by three sensors (gamma-ray spectrometer, RS 230;
susceptibilimeter KT10, Terraplus; and conductivimeter, EM38 Geonics) at
75 points and analyzed the data. The models with the best performance
(R2 0.48, 0.36, 0.44, 0.36, 0.25, and 0.31) varied for clay, sand,
Fe2O3, TiO2, SiO2, and cation exchange capacity
prediction, respectively. Modeling with the selection of covariates at three
phases (variance close to zero, removal by correction, and removal by
importance) was adequate to increase the parsimony. The results were
validated using the method “nested leave-one-out cross-validation”. The
prediction of soil attributes by machine learning algorithms yielded
adequate values for field-collected data, without any sample preparation,
for most of the tested predictors (R2 values ranging from 0.20 to
0.50). Also, the use of four regression algorithms proved to be important
since at least one of the predictors used one of the tested algorithms. The
performance values of the best algorithms for each predictor were higher
than those obtained with the use of a mean value for the entire area
comparing the values of root mean square error (RMSE) and mean absolute
error (MAE). The best combination of sensors that reached the highest model
performance was that of the gamma-ray spectrometer and the
susceptibilimeter. The most important variables for most
predictions were parent material,
digital elevation, standardized height, and magnetic susceptibility. We concluded that soil attributes can be efficiently modeled
by geophysical data using machine learning techniques and geophysical sensor
combinations. This approach can facilitate future soil mapping in a more
time-efficient and environmentally friendly manner.
•Weathering and pedogenesis on landscape affect distribution radionuclides.•Lithology strongly affect radionuclides content on less evolved soil classes.•Landscape evolution can affect distribution ...of radionuclides.•Soil texture and other attributes can be inferred by radionuclides content.•Gamma-ray spectrometry can be used in digital soil mapping.
Geophysical methods, such as gamma-ray spectrometry, have great potential to enhance knowledge of the pedosphere (pedogenesis, pedogeochemistry and pedogeomorphology), helping to predict tropical soil attributes. We applied proximal gamma-ray spectrometry to evaluate tropical landscape dynamics, pedogenesis and spatial distribution of radionuclides and selected soil attributes. This study was carried out in southeast Brazil, where 79 soil samples (0–20 cm) along transects were collected to perform physical-chemical analysis coupled with collection of surface gamma-ray spectrometric data, allowing the detection of the radionuclides uranium (U238), thorium (Th232), and potassium (K40). Additionally, we analyzed soils in four toposequences, with varying lithology, relief, and hydrology. The radionuclide concentrations in soils showed a direct relationship with the parent material composition, either rocks or sediments. Weathering degree and the geochemical behavior of each radionuclide determines its permanence or removal in soils. Denudation processes along the toposequences also influence the distribution of radionuclides. On one hand, the radionuclide contents of mature, weathered soils are closely associated with advanced pedogenesis, with a higher clay contents and argic horizon formation, whereas in younger, less weathered soils the parent material exerts a greater influence than pedogenesis. Uranium decreased with altitude, and showed greater mobility compared with thorium. Thorium presented a higher correlation with clay content, and the opposite with sand content. Potassium detection increased with soil cation exchange capacity (CEC) and clay content. Gamma-spectrometry detected significant variations in some segments along the toposequences, undetectable by conventional soil survey techniques. This may indicate changes from one soil class to another or the continuity of a particular soil class, demonstrating the potential of this tool in digital soil mapping, pedogeochemical, pedogeomorphological and pedogenesis processes studies.
Display omitted
•Combined use of geophysical sensors in modeling pedogenetic processes.•Use of Nested Cross Validation as a form of external validation.•Use soil, lithology, relief and satellite ...image to determine ideal number of clusters.•Geophysical data, machine learning and clustering identified different pedogenesis rates.•Use of clusters in modeling processes.
Pedogenetic processes such as ferralitization and argilluviation control various soil attributes. Understanding the intensities of pedogenesis can provide answers for several fields of environmental studies, including soil science and the geosciences. Recently, new geotechnologies such as geophysics applied to soil science and machine learning algorithms have proven to be a potential tool in pedosphere studies. In this research, we performed component principal analyses and determined the ideal number of clusters based on geophysical soil data and satellite images. Then, we used the ideal number of clusters, and ferralitization and argilluviation indices, as input data in five modeling (prediction and spatialization) algorithms to infer different ferralitization and argilluviation intensities in soils formed from the same soil parent material. The results showed that avNNet had the best model performance for modeling the clusters showing that the ideal number of clusters was three. The variables which contributed the most to the modeling were the solar diffuse radiation, topographic wetness index, and digital elevation model. The model’s specificity was greater than its sensitivity. The areas over diabase and Nitisols in the east of the study area presented greater ferralitization rates than diabase and Nitisols over western areas. On the other hand, the areas over siltite and Lixisols in the east presented greater argilluviation rates than metamorphosed siltite/siltite and Lixisols over western areas. The relief and topographic position strongly affected the evaluated pedogenetic processes, since they controlled the hydric dynamics in the area. The geophysical variables were related to soil attributes and pedogenesis, which contributed to modeling and clusterization processes.
•Removals of iron forms and organic matter alter absorption features in the mid-IR.•The addition of water have stronger influence in sandy than in clay soil.•High correlation values were observed for ...Kt, Gb, SiO2 and Fe2O3 with ASTER bands.•New mid-IR regions have been proposed for satellite remote sensing.
The soil mineralogical constitution directly influences its chemical, physical and hydraulic characteristics. Although very important, it is still rarely used for decision-making in agriculture, mainly due to the complexity and cost of standard analyzes. In this sense, the middle infrared spectroscopy (mid-IR, 4000 to 400 cm−1) has great potential to obtain soil mineralogical information quickly and accurately. Nevertheless, some soil constituents can severely influence the spectra and produce misinterpretations. In this research, we aim to detect changes in the mid-IR spectra caused by water, iron forms and organic matter (OM), and to relate soil attributes to laboratory spectra and remote sensing simulated spectral bands. The research area is located in São Paulo State, Brazil, where seventeen soil samples were collected. The reflectance intensities, shapes and absorption features of the mid-IR spectra before and after the removal of OM and iron forms and the addition of water were described. Soil attributes, such as kaolinite, gibbsite, 2:1 minerals among others were correlated with the mid-IR spectra and simulated ASTER spectral bands by Pearson's analysis, to verify its potential on mineralogical evaluation. The description of mid-IR revealed that the removal of the OM from the soil samples decreased the reflectance intensities between 4000 and 2000 cm−1. Iron forms mainly influence the 3250 – 1200 cm−1 spectral range and mask the spectral features of other minerals as well. The addition of water masked several absorption features and decreased the reflectance intensities from 3700 to 2700 cm−1. High correlation coefficients were obtained between soil attributes and ASTER simulated spectral bands, which allowed the selection of potential spectral regions for future satellite sensors: 2760 – 2500 cm−1 (3600 – 4000 nm), 2150 – 1875 cm−1 (4600 – 5300 nm), and 840 – 740 cm−1 (11900 – 3500 nm).
The expansion of agricultural frontiers has increased the use of sandy soils as cropping areas; however, the water‐retention capacity (WRC) of the soil is a crucial factor. Thus, fast and accurate ...methodologies, such as spectroscopy, have gained importance to identify physical‐hydric and other attributes related to the WRC. Some theories indicate the micromorphometry of particles is an important factor in sandy soils. This study (I) investigated the WRC of sandy soils from distinct geologic formations and the influence of particle micromorphometry on the WRC; (II) evaluated the sensitivity of Vis–NIR and MIR spectra to characterise physical‐hydric attributes and particle micromorphometry; and (III) quantified physical‐hydric attributes of sandy soils from their physical attributes, particle micromorphometric indices, and the visible‐near infrared‐mid infrared (Vis–NIR‐MIR) spectra. We determined the values of field capacity (FC), permanent wilting point (PWP), available water (AW), clay, sand, silt, sand fractions, and OM as well as the micromorphometric indices and Vis–NIR and MIR reflectance of 29 sandy soil profiles. Soil attributes were correlated with each other and with the Vis–NIR and MIR spectra to understand their relationship with the WRC. The correlation and the principal component analyses (PCA) were performed to understand the relationship between Vis–NIR and MIR spectra and sand grain micromorphometry. Mathematical models were developed to estimate FC, PWP, and AW using soil attributes, micromorphometry, Vis–NIR, and MIR as predictor variables. Roughness of sand particle greatly influenced the WRC of sandy soils. The MIR region showed high sensitivity to identify different WRC patterns and particle micromorphometry in sandy soils. The Vis–NIR and MIR spectra presented high accuracy to predict FC and PWP of sandy soils. The mathematical models that used soil attributes + micromorphometry presented excellent calibration parameters; however, they did not show accuracy in the validation set.
Highlights
The influence of particle roughness on water retention of sandy soils was proven.
Roughness had a greater influence on water retention than texture proportions for highly sandy soils.
MIR spectra correlate strongly with the roughness of the very fine sand fraction.
MIR spectra and roughness can estimate sandy soils water retention with high accuracy.
The aim of the current research was to examine the effectiveness of Vis-NIR-SWIR (visible, near-infrared, and shortwave infrared spectroscopy: 350–2500 nm), MIR (mid-infrared spectroscopy: ...4000–400 cm−1), and pXRF (portable x-ray fluorescence) to characterize and estimate clay, sand, silt, calcium carbonate (CaCO3), gypsum (CaSO4.2H2O), soil organic carbon (SOC) contents and electrical conductivity (EC) in Afzar district, Fars province, southern Iran. The study area was divided into non-saline (A) and saline (B) regions, and then a total of 300 soil samples from these areas were collected for laboratory analysis. The partial least-squares regression (PLSR) method was used to predict soil properties from Vis-NIR-SWIR, mid-IR, and pXRF spectra. In general, Vis-NIR-SWIR showed better results for predicting soil texture (R2 for area A: clay = 0.78, sand = 0.80 and silt = 0.69; R2 for area B: clay = 0.63, sand = 0.67 and silt = 0.47) in the arid regions than mid-IR (R2 for area A: clay = 0.71, sand = 0.75 and silt = 0.63; R2 for area B: clay = 0.61, sand = 0.57 and silt = 0.37), and pXRF (R2 for area A: clay = 0.49, sand = 0.60 and silt = 0.50; R2 for area B: clay = 0.41, sand = 0.41 and silt = 0.23). Based on the RPI index, three approaches alone could successfully predict CaCO3 in areas A and B (R2 > 0.75 and RPIQ >2.02). The model prediction results for gypsum showed that although the value of the determination coefficient is high (R2 > 0.95), but according to the low value of the RPIQ index (RPIQ<1.26), the model prediction performance is untrusted, and this is probably due to unsuitable distribution of gypsum in this area. The results showed that the pXRF technique alone could not predict SOC in these areas (R2 < 0.27 and RPIQ <1.08), and it is probably better to use it in combination with Vis-NIR-SWIR and mid-IR data. pXRF for predicting soil salinity in areas A and B was acceptable and successful, respectively. Also, the results showed that the Vis-NIR-SWIR and mid-IR ranges could not predict salinity in these regions and should be used in combination with pXRF data. In conclusion, the Vis-NIR-SWIR, mid-IR, and pXRF spectroscopy are presented here as useful and reliable tools for direct applications in pedology, particularly in the digital mapping of soil surface properties.
•The effectiveness of Vis-NIR-SWIR, and MIR were examined to estimate some soil properties.•Vis-NIR-SWIR showed better results for soil texture than mid-IR and PXRF.•Each technique alone could successfully predict CaCO3 in the study area•The results showed that the PXRF technique alone is not able to predict SOC in these areas.•Vis-NIR-SWIR, mid-IR and PXRF reflectance spectroscopy are presented here as useful and reliable tools
The GeoWEPP model has estimated water and soil losses caused by erosion at the watershed level in different parts of the world. However, this model was developed and its parameters have been adjusted ...for temperate climates, which are different from tropical climates such as those found in Brazil. Our study evaluated the performance of the GeoWEPP model in estimating soil erosion in three micro-watersheds in the Cerrado (i.e., savannah) of southeastern Mato Grosso state, Brazil. Major land uses modeled were soybean and corn cultivation, traditional pasture, and native vegetation. Input parameters for the GeoWEPP model involved climate, soil, land use and management, and topography. GeoWEPP was calibrated with input parameters for soil erodibility specified as interrill and rill soil erosion, soil critical shear stress, and saturated hydraulic conductivity obtained experimentally and estimated by internal routine equations of the GeoWEPP model. Soil losses observed in micro-watersheds with agriculture, pasture, and native vegetation were 0.11, 0.06, and 0.10 metric tons per hectare per year, respectively. GeoWEPP best modeled soil erosion for native vegetation and pasture, while over-estimating that for crops. Surface runoff was best modeled for crops versus native vegetation and pasture. The GeoWEPP model performed better when using soil erodibility input parameters.
ABSTRACT The method of applying fertiliser to the soil affects grain productivity. The aim of this study was to determine the dynamics of phosphorus, potassium and sulphur in the arable layer of the ...soil during the third year of agricultural production, as a function of the method of application. The study took place in the district of Diamantino, Mato Grosso, in an area of soybean-maize succession under minimum tillage. Five treatments were tested in a randomised block design: no phosphorus or potassium (P0-K0); broadcast phosphorus and in-furrow potassium (Pb-Kf); in-furrow phosphorus and potassium (Pf-Kf); in-furrow phosphorus and broadcast potassium (Pf-Kb); broadcast phosphorus and potassium (Pb-Kb). When the soybeans reached physiological maturity, soil samples were collected from the 0-0.05, 0.05-0.10, 0.10-0.15 and 0.15-0.20 m layers for analysis of the pH, and P, K, S and organic matter content. The soil attributes were submitted to analysis of variance and Tukey’s test at 5%, and the nutrient dynamics in the soil profile were analysed by regression. The in-furrow application of phosphorus (Pf-Kb and Pf-Kf) improved P availability in the 0-0.05 m layer by 69.12% and 46.54% compared to the Pb-Kb treatment, by 73.93% and 50.71% compared to the Pb-Kf treatment, and by 136.77% and 105.16% compared to the P0-K0 treatment. The in-furrow application of phosphate fertiliser proved to be more effective in terms of phosphorus availability in the soil. The residual effect of broadcasting phosphate fertiliser on phosphorus availability in the soil can be seen from the third year of production onwards.
Despite occupying a large area of the globe and being the next agricultural frontier, sandy soils are seldom explored in scientific studies. Considering the high capacity of remote sensing in soil ...characterization, this work aimed to: (i) characterize sandy soils’ profiles from proximal sensing; (ii) assess the ability of visible, near, and short-wave infrared (Vis-NIR-SWIR) as well as mid-infrared (MIR) spectroscopy to distinguish soil classes of highly sandy content; (iii) quantify physical and chemical attributes of sandy soil profiles from Vis-NIR-SWIR and MIR spectroscopy as well as X-ray fluorescence (pXRF). Samples were described and collected from 29 sandy soil profiles. The 127 samples went under Vis-NIR-SWIR and MIR spectroscopy, X-ray fluorescence, and chemical and physical analyses. The spectra were analyzed based on “Morphological Interpretation of Reflectance Spectrum” (MIRS), Principal Components Analysis (PCA), and cluster methodology to characterize soils. The integration of different information obtained by remote sensors, such as Vis-NIR-SWIR, MIR, and Portable X-ray Fluorescence (pXRF), allows for pedologically complex characterizations and conclusions in a short period and with low investment in analysis and reagents. The application of MIRS concepts in the VNS spectra of sandy soils showed high potential for distinguishing pedological classes of sandy soils. The MIR spectra did not show distinct patterns in the general shapes of the curves and reflectance intensities between sandy soil classes. However, even so, this region showed potential for identifying mineralogical constitution, texture, and OM contents, assuming high importance for the complementation of soil pedometric characterizations using VNS spectroscopy. The VNS and MIR data, combined or isolated, showed excellent predictive performance for the estimation of sandy soil attributes (R2 > 0.8). Sandy soil color indices, which are very important for soil classification, can be predicted with excellent accuracy (R2 from 0.74 to 0.99) using VNS spectroscopy or the combination of VNS + MIR.
Abstract The objective of this work was to determine pedotransfer functions to predict water retention at the -33 and -1500 kPa matric potentials of soils under Cerrado, in the south of the state of ...Mato Grosso, Brazil. Samples (n=156) were collected for model calibration (n=124) and validation (n=32). A stepwise multiple regression was used to determine pedotransfer functions. Willmott’s index of agreement, root-mean-square error, Pearson’s correlation coefficient, and the confidence index were used to evaluate the performance of the functions and to compare them with those described in the literature. The most efficient variables to estimate water retention were: microporosity, total sand, and clay at -33 kPa; and total sand, silt and clay at -1500 kPa. The regional pedotransfer functions explained more than 94% of water retention variance in the studied soils. The reliability of the functions to predict water retention increased, at -33 kPa, with the use of the structural property microporosity and, at -1500 kPa, with the use of granulometric parameters. The studied regional pedotransfer functions predict water retention at the -33 and -1500 kPa matric potentials of soils under Cerrado, in southern Mato Grosso, better than the functions described in the literature.
Resumo O objetivo deste trabalho foi determinar funções de pedotransferência para a predição da retenção de água nos potenciais matriciais -33 e -1500 kPa, em solos sob Cerrado, no sul do estado de Mato Grosso, Brasil. Amostras (n=156) foram coletadas para a calibração (n=124) e validação (n=32) dos modelos. Utilizou-se uma regressão múltipla passo a passo, para determinar as funções de pedotransferência. Utilizaram-se o índice de concordância de Willmott, a raiz do erro quadrático médio, o coeficiente de correlação de Pearson e o índice de confiança, para avaliar o desempenho das funções e compará-las com as descritas na literatura. As variáveis mais eficientes para estimar a retenção de água foram: microporosidade, areia total e argila a -33 kPa; e areia total, silte e argila a -1500 kPa. As funções de pedotransferência regionais explicaram mais de 94% da variância da retenção de água nos solos estudados. A confiabilidade da predição da retenção de água, -33 kPa, aumentou com o uso da propriedade estrutural microporosidade e, a -1500 kPa, com o uso de parâmetros granulométricos. Funções de pedotransferência regionais predizem melhor a retenção de água nos potenciais matriciais -33 e -1500 kPa, em solos sob Cerrado, no sul de Mato Grosso, do que as funções descritas na literatura.