Soil Spectral Library Research Articles

Soil Color and Mineralogy Mapping Using Proximal and Remote Sensing in Midwest Brazil

Soil color and mineralogy are used as diagnostic criteria to distinguish different soil types. In the literature, 350–2500 nm spectra were successfully used to predict soil color and mineralogy, but these attributes currently are not mapped for most Brazilian soils. In this paper, we provided the first large-extent maps with 30 m resolution of soil color and mineralogy at three depth intervals for 850,000 km2 of Midwest Brazil. We obtained soil 350–2500 nm spectra from 1397 sites of the Brazilian Soil Spectral Library at 0–20 cm, 20–60, and 60–100 cm depths. Spectra was used to derive Munsell hue, value, and chroma, and also second derivative spectra of the Kubelka–Munk function, where key spectral bands were identified and their amplitude measured for mineral quantification. Landsat composites of topsoil and vegetation reflectance, together with relief and climate data, were used as covariates to predict Munsell color and Fe–Al oxides, and 1:1 and 2:1 clay minerals of topsoil and subsoil. We used random forest for soil modeling and 10-fold cross-validation. Soil spectra and remote sensing data accurately mapped color and mineralogy at topsoil and subsoil in Midwest Brazil. Hematite showed high prediction accuracy (R2 > 0.71), followed by Munsell value and hue. Satellite topsoil reflectance at blue spectral region was the most relevant predictor (25% global importance) for soil color and mineralogy. Our maps were consistent with pedological expert knowledge, legacy soil observations, and legacy soil class map of the study region.

Improvement in spectral library-based quantification of soil properties using representative spiking and local calibration – The case of soil inorganic carbon prediction by mid-infrared spectroscopy

Mid-infrared reflectance spectroscopy (MIRS) is time- and cost-effective. It was used for quantifying soil inorganic carbon (SIC) concentration in France based on a national library, and performances were evaluated on an independent regional set. Our objective was to improve the accuracy of MIRS predictions based on common multivariate regression, through spiking (enrichment of the national library with some representative target samples) with possible extra-weighting (replication of spiking samples) and local calibration (only using calibration samples that are spectral neighbours of each target samples), which have not been fully explored yet, in combination especially.Global (i.e. common) calibration yielded accurate prediction (standard error of prediction, SEP, was ≈5 g kg−1), which could be improved when the library was completed with spiking samples (optimally 10samples extra-weighted 40times; SEP = 3.3 g kg−1). Using spiking samples only (without the library) yielded slightly less accurate results (SEP = 3.6 g kg−1). Prediction was more accurate using local calibration without spiking, but on a validation set that was reduced because some validation samples lacked calibration neighbours (SEP = 2.5–2.7 g kg−1). Local calibration with spiking (optimally 10samples without extra-weight) yielded somewhat less accurate prediction but for the full validation set when few calibration neighbours were required (SEP = 2.7 g kg−1), or higher accuracy on the reduced validation set when many neighbours were required (SEP = 2.3 g kg−1).These accurate predictions demonstrated the usefulness of representative spiking and local calibration for rendering large soil spectral libraries fully operational, while extra-weighting had no additional benefit. Along with more exhaustive spectral libraries, this paves the way for extensive use of MIRS for SIC determination.

Simultaneous prediction of soil properties from VNIR-SWIR spectra using a localized multi-channel 1-D convolutional neural network

The use of visible near-infrared and shortwave-infrared (VNIR-SWIR) diffuse reflectance spectroscopy for the estimation of soil properties is increasingly maturing with large-scale soil spectral libraries (SSLs) of laboratory spectra developed across the globe. Such an SSL is the publicly available LUCAS topsoil database with approximately 20,000 soil samples encompassing 23 countries of the European Union. A wide variety of machine learning tools have been applied to the LUCAS SSL to predict some of the soil samples’ physicochemical properties with different degrees of accuracy. In this paper, we developed and examined the use of a novel one-dimensional convolutional neural network (CNN) to simultaneously predict ten physicochemical properties of the LUCAS SSL. Leveraging on the use of multiple-input channels it uses as model inputs the absorbance spectra along with some pre-processed spectra developed using standard techniques. Moreover, it exploits the use of local spectral neighborhoods to perform an adaptive error-correction mechanism. This novel localized multi-channel 1-D CNN was applied to all the available physicochemical properties of the LUCAS SSL and was statistically compared with the current state-of-the-art where it was shown to statistically outperform its counterparts, as well as with other CNNs where it exhibited the best performance. In particular, for the mineral soil samples, the RMSE for the Clay content was 4.80% (R2 0.86), for soil organic carbon the RMSE was 10.96 g kg−1 (R2 0.86), while for total nitrogen the RMSE was 0.66 g kg−1 (R2 0.83).

A new model for intra- and inter-institutional soil data sharing

Abstract. Data sharing and collaboration are critical to solving large-scale problems. The prevailing soil data-sharing model is based on different groups sending their data to a lead party. This model is of a centralised nature and, consequently, results in the participants ceding control and governance over their data to the lead party. Here we explore the use of a distributed ledger (blockchain) to solve the aforementioned issues. We explain what a blockchain is and some of its characteristics to then describe some features of a blockchain that make it an interesting candidate for an inter-institutional database. Finally, we describe the potential use case of developing a global soil spectral library with multiple, independent international institutions constituting the network.

VisNIR integrated multi-sensing penetrometer for in situ high-resolution vertical soil sensing

An in situ penetrometer system that can measure profile soil properties rapidly, cost-effectively, and at high vertical resolution would benefit the soil science and agriculture communities. A visible and near infrared (VisNIR) integrated multi-sensing penetrometer system was developed to automatically measure in situ soil VisNIR reflectance spectra, penetration resistance, and insertion depth along a soil profile. The system was tested in 11 agricultural fields in Nebraska, Illinois, Iowa, and South Dakota of the U.S. An independent soil VisNIR spectral library was used to build calibration models for soil property prediction. External Parameter Orthogonalization (EPO) was used to correct for the field intactness of in situ VisNIR spectra. The results showed that EPO was effective in correcting for the spectral disparity between in situ and dry-ground VisNIR spectra. The EPO correction showed an improvement of prediction accuracy of soil total carbon (R2 and RMSE improved from 0.29 and 3.06 % to 0.5 and 0.79 %, respectively) and total nitrogen (R2 and RMSE improved from 0.51 and 0.36 % to 0.62 and 0.06 %, respectively). The system also predicted soil bulk density with an RMSE of 0.12 g cm−3 and R2 of 0.80. It is concluded that the VisNIR multi-sensing penetrometer, along with the use of external soil spectral libraries and the spectral correction algorithm EPO, can lead to a rapid, robust and cost-effective system for in situ high resolution vertical soil sensing.

Identifying and mapping terrons in Denmark

Spatial assessment of terroir is creating a new possibility for enhancement of high quality agro-food product and to minimize negative environmental effects such as soil degradation and associated risks. The classification and mapping of particular terroir units could be a competitive marketing tool with a major impact on farmers’ incomes. For this purpose, Carré and McBratney (2005) proposed the terron concept to establish combined soil and landscape entities as the first investigative step to identify terroirs. The main objective of the present work was to assemble various environmental factors (i.e. soil, terrain and climate), to identify and then to map terrons in Denmark. First, for representing soil factors, a national soil spectral library was utilized to measure taxonomic distances between 34 Danish reference soil profiles and the Danish national soil profile database (586 soil profiles). Second, the terrain and climate factors for each soil profile location were then compiled as represented by relative slope position, valley depth, valley bottom flatness, vertical distance to the channel network, number of frost days, annual number of growing days, global solar radiation, and precipitation. Third, nine Danish terron classes were established by fuzzy c-means clustering based on an integrated matrix including all soil, terrain and climate factors whereby each terron class is characterized by soil, terrain and climate as a whole entity. Finally, the spatial distribution of Danish terrons was mapped using Cubist regression rules. The results were compared with a soil map derived from the same profile database. We concluded that the map of terrons described natural environment quantitatively and formally in terms of soil, landscape and climatic information better than just a soil class or soil attribute map. Further investigations are needed to discover whether the terron classes give better predictions of landscape-dynamic processes and allow better management options than soil alone. This study also demonstrated several advantages of using soil spectral data and ancillary data to identify and map terrons. The next step will be to validate the terron map by incorporating crop yield data and social factors to delineate natural Danish terroir units.

Mapping at 30 m Resolution of Soil Attributes at Multiple Depths in Midwest Brazil

The Midwest region in Brazil has the largest and most recent agricultural frontier in the country where there is no currently detailed soil information to support the agricultural intensification. Producing large-extent digital soil maps demands a huge volume of data and high computing capacity. This paper proposed mapping surface and subsurface key soil attributes with 30 m-resolution in a large area of Midwest Brazil. These soil maps at multiple depth increments will provide adequate information to guide land use throughout the region. The study area comprises about 851,000 km2 in the Cerrado biome (savannah) in the Brazilian Midwest. We used soil data from 7908 sites of the Brazilian Soil Spectral Library and 231 of the Free Brazilian Repository for Open Soil Data. We selected nine key soil attributes for mapping and aggregated them into three depth intervals: 0–20, 20–60 and 60–100 cm. A total of 33 soil predictors were prepared using Google Earth Engine (GEE), such as climate and geologic features with 1 km-resolution, terrain and two new covariates with 30 m-resolution, based on satellite measurements of the topsoil reflectance and the seasonal variability in vegetation spectra. The scorpan model was adopted for mapping of soil variables using random forest regression (RF). We used the model-based optimization by tuning RF hyperparameters and calculated the scaled permutation importance of covariates in R software. Our results were promising, with a satisfactory model performance for physical and chemical attributes at all depth intervals. Elevation, climate and topsoil reflectance were the most important covariates in predicting sand, clay and silt. In general, for predicting soil chemical attributes, climatic variables, elevation and vegetation reflectance provided to be the most important of predictive components, while for organic matter it was a combination of climatic dynamics and reflectance bands from vegetation and topsoil. The multiple depth maps showed that soil attributes largely varied across the study area, from clayey to sandy, suggesting that less than 44% of the studied soils had good natural fertility. We concluded that key soil attributes from multiple depth increments can be mapped using Earth observations data and machine learning methods with good performance.

Predicting soil permanganate oxidizable carbon (POXC) by coupling DRIFT spectroscopy and artificial neural networks (ANN)

Infrared spectroscopy has transformed soil property quantification by enabling low-cost, high-throughput analysis of soils, enabling mapping and monitoring of this non-renewable resource. However, less evaluated are newly emerging indicators of soil health. Furthermore, as soil spectral libraries expand in size, commonly employed linear models such as partial least squares regression (PLSR) may be challenged by the number and diversity of spectra. Artificial neural networks (ANN) are an emerging deep learning approach that can offer advantages in quantification of soil properties by utilizing non-linear relationships among spectra and soil components. We compared ANN versus PLSR models for predicting an increasingly used soil health indicator, permanganate oxidizable C (POXC), as well as more routinely predicted soil variables (e.g., clay, soil organic C [SOC]), across a gradient of soil organic matter furnished by a deforestation chronosequence in Kenya (n = 144). Candidate ANN architectures were first methodologically evaluated and described to identify best-practices for the application of ANN to soil spectroscopy. Predictions by the resulting ANN relative to PLSR were similar or slightly improved for routinely measured variables that represent soil organic matter (SOC, C:N) and physical properties (clay, silt, sand, bulk density). The accuracy of POXC predictions were similar for ANN (RMSE 102 mg kg−1) and PLSR (RMSE 106 mg kg−1). However, models drew on shared but also distinct wavenumbers, indicating differential use of information in soil infrared spectra by non-linear versus linear chemometric models. Even in relatively small spectral datasets of similar soil types expected to favor PLSR, ANN shows comparable predictive performance. To help guide future applications of ANN in soil spectroscopy, we propose a systematic procedure to select ANN model hyperparameters.

Impacts of land use and topography on soil organic carbon in a Mediterranean landscape (north-western Tunisia)

Abstract. This study evaluates the impact of land use and topographic features (slope and aspect) on soil organic carbon (SOC) within the Wadi Beja watershed in north-western Tunisia. A soil spectral library was set up to assess the variation in the SOC for 1440 soil samples from four land use types (field crops, permanent crops, forest, and grazing land), three slope categories (flat, moderate, and steep) and two aspects (north- and south-facing). For field crops, only one factor – slope – significantly affected SOC, with SOC content in north-facing areas appearing to be higher in flat areas (0.75 %) than in hilly areas (0.51 %). However, in south-facing areas, SOC content was also higher in flat areas (0.74 %) than in hilly areas (0.50 %). For permanent crops, which were inter-planted with field crops, the slope significantly affected SOC content, which improved to 0.97 % in flat north-facing and 0.96 % in flat south-facing areas, scoring higher than hilly south- and north-facing areas (0.79 %). In the grazing land use system, both of the investigated factors – aspect and slope – significantly affected the SOC content, which was significantly higher in flat areas (north-facing: 0.84 %, south-facing: 0.77 %) than in hilly areas (north-facing: 0.61 %, south-facing: 0.56 %). For the forest, none of the factors had a significant effect on SOC content, which was higher in flat areas (north-facing: 1.15 %, south-facing: 1.14 %) than in steep areas (1.09 % in north-facing areas and 1.07 % in south-facing areas). This study highlights the ability of visible and near-infrared (VNIR) spectroscopy to quantify C in diverse soils collected over a large diverse geographic area in order to indicate that calibrations are feasible, and therefore, assessing the variation of SOC content under land use and topographic features (slope and aspect) will result in better sustainable land management planning.

Preparing a soil spectral library using the Internal Soil Standard (ISS) method: Influence of extreme different humidity laboratory conditions

Extensive efforts are currently being devoted to the establishment of soil spectral libraries on regional, national, continent-wide and global domains. In particular, a new development goes in the direction of global harmonized soil spectral databases that shall be acquired following common standards and procedures so that they can be merged with other soil spectral libraries. The choice of standards is important, especially with respect to their behaviors when laboratory conditions, such as humidity, change. In this study, we test the application and robustness of the Internal Soil Standard (ISS) spectral re-alignment procedure on an extended number of soil samples acquired at different laboratories. In particular, a focus is placed on the comparison between two standards Lucky Bay (LB) and Wylie Bay (WB), for their performance with different humidity laboratory conditions. LB and WB are almost pure quartz sands from Australia. For this, 71 soil samples from Israel with different mineralogical background and variable soil organic matter contents are scanned at two laboratories. The scanning took place with different spectral measurement protocols and extreme diverse conditions in terms of laboratory humidity and moist. All scans are completed by the repeated scans of the two Australian white sands through all scan batches to harmonize the spectral measurements. Our results show that the ISS minimizes the visual spectral variation, aligns the minor and extreme systematic changes of the protocols, and makes them more stable. Furthermore, the LB and WB standards do not exhibit equal performance regarding to relatively dry and humid conditions. The WB standard provides more stable and satisfactory results in humid condition. However, the high performance of the LB sample in spectral correction is still observed, particularly in laboratories with lower moist. Accordingly, these analyses suggest that in the ISS re-alignment procedure, the WB sample is more reliable to be used in humid laboratory condition.

The Brazilian Soil Spectral Library (BSSL): A general view, application and challenges

The present study was developed in a joint partnership with the Brazilian pedometrics community to standardize and evaluate spectra within the 350–2500 nm range of Brazilian soils. The Brazilian Soil Spectral Library (BSSL) began in 1995, creating a protocol to gather soil samples from different locations in Brazil. The BSSL reached 39,284 soil samples from 65 contributors representing 41 institutions from all 26 states. Through the BSSL spectra database, it was possible to estimate important soil attributes, such as clay, sand, soil organic carbon, cation exchange capacity, pH and base saturation, resulting in differences among the multi-scale models taking Brazil (overall), regional and state scale. In general, spectral descriptive and quantitative behavior indicated important relationship with physical, chemical and mineralogical properties. Statistical analyses showed that six basic patterns of spectral signatures represent the Brazilian soils types and that environmental conditions explain the differences in spectra. This study demonstrates that spectroscopy analyses along with the establishment of soil spectral libraries are a powerful technique for providing information on a national and regional levels. We also developed an interactive online platform showing soil sample locations and their contributors. As soil spectroscopy is considered a fast, simple, accurate and nondestructive analytical procedure, its application may be integrated with wet analysis as an alternative to support the sustainable management of soils.

Strategies for the efficient estimation of soil organic carbon at the field scale with vis-NIR spectroscopy: Spectral libraries and spiking vs. local calibrations

Soil spectroscopy in the visible-to-near-infrared (vis-NIR) range is a cost-effective alternative analysis technique to determine soil organic carbon (SOC). The development and provision of large-scale soil spectral libraries (SSLs) further facilitates the application of vis-NIR spectroscopy for the rapid assessment of SOC. However, optimal strategies to apply SOC calibrations from SSLs to independent field sites have yet to be established.We tested the predictive ability of SOC calibrations based on three external SSLs at the national, regional and field scale by applying them to two field sites in Germany. The national-scale SSL was comprised as a subset of the European LUCAS 2009 topsoil database. This subset was further classified into a randomly selected subset and target site-specific subsets based on similarity of spectral characteristics and soil parent material. A regional- and a field-scale legacy dataset were additionally used to predict SOC at the two field sites and to compare the results with the performance of the LUCAS based models. SSL-based predictive models adapted to the characteristics of the target sites by means of spiking were evaluated against purely local calibrations.Models calibrated with spectra from the LUCAS library and the regional-scale dataset predicted the SOC contents of the target field sites generally poorly (0.45 ≤ RPD ≤ 2.08), largely as a result of biased estimates. Spiking the models with only a few (~15) samples from the target sites reduced prediction bias drastically and thus yielded markedly improved SOC estimates for nearly all redeveloped models (1.30 ≤ RPD ≤ 3.69). Spiking the models based on the field-scale SSL with 15 samples produced better results than the spiked larger national and regional calibration sets, with RPD values of 5.66 and 4.14 for both target sites.Our results suggest that universal calibrations based exclusively on library spectra of larger scale are insufficient for accurate SOC assessments at the local scale even with pedogenetically or spectrally adapted calibration subsets. Spiking the vis-NIR models based on SSLs with a small number of target site samples allows a successful transfer of SOC calibrations, but does not necessarily yield more accurate predictions than local models developed exclusively with the spiking samples or calibrations based on field-scale SSL with similar characteristics, which may be preferable for model development if available.

A Spectral Assignment-Oriented Approach to Improve Interpretability and Accuracy of Proxy Spectral-Based Models

In modeling chemical attributes using hyperspectral data, nonlinear relationships between the predictor and the response are frequent. The common nonlinear modeling techniques improve prediction accuracy but suffer from low interpretability of the models. In this paper, we demonstrate a new multivariate modeling method, denoted as spectral assignment-oriented partial least squares (SAO-PLS), which is designed to provide a nonlinear modeling solution with strong interpretability products. The need for this approach is apparent when a given sample population consists of different spectral features for different levels of the response. Accordingly, the suggested SAO-PLS algorithm segments the data in an optimal location on the response distribution by maximizing the difference in spectral assignments between two clusters. SAO-PLS is applied here to two test cases with different characteristics: 1) an established data set containing airborne hyperspectral data of asphaltic roads, merged with in situ measured dynamic friction values captured using a standardized method and 2) a soil spectral library, spectrally measured with an analytical spectral device spectrometer, to which organic carbon measurements were applied. Our results demonstrate the superiority of SAO-PLS over partial least-squares regression for both model accuracy and interpretability, providing a deeper understanding of the underlying processes.

A Framework for Global Characterization of Soil Properties Using Repeat Hyperspectral Satellite Data

Imaging spectroscopy offers the potential to quantify the soil properties over large areas based on its reflectance spectra. Soils are heterogeneous mixtures of minerals, air, water, and organic matter leading to complex manifestations of reflectance in the different parts of the visible-shortwave infrared spectra. Due to this complexity, data-driven modeling approaches are found to be most suitable for characterizing the relationships between soil spectra and the corresponding soil properties. Proposed spaceborne hyperspectral missions, such as Hyperspectral Infrared Imager, offer the possibility of repeating global spectral measurements in a 16- to 20-day revisit period. Soil attributes on the landscape vary at different rates. In particular, the soil textural attributes (percentage of sand, silt, and clay) may be assumed to remain invariant compared to chemical constituents during multiple consecutive 16- to 20-day satellite revisit period. We present a theoretical retrieval framework for assimilating repeat spaceborne soil spectral measurements into a previously developed lasso algorithm-based ensemble modeling framework for the global-scale characterization of soil textural attributes. The repeat spectral assimilation with each overpass of the satellite leads to the development of an enriched “dynamic soil spectral library” which spatially propagates the improvement in the characterization of soil textural properties globally, given the uncertain variations in other auxiliary factors, such as moisture and organic matter, affecting soil reflectance.

An evolutionary fuzzy rule-based system applied to the prediction of soil organic carbon from soil spectral libraries

The recent efforts for obtaining vast soil spectral libraries covering a significant part of the spatial and compositional variability of soils have underscored the need for accurate and interpretable models. Herein, the application of an evolutionary Fuzzy Rule-based System (FRBS) named DECO3RUM (Differential Evolution based Cooperative and Competing learning of Compact Rule-based Models) for the prediction of soil properties from visible, near-infrared, and shortwave-infrared (VNIR–SWIR) laboratory spectral data obtained from the LUCAS topsoil database is investigated. FRBSs model the input–output relation with fuzzy logic statements, offering an enhanced interpretability degree for the experts over classical rule-based systems and other black box models. The proposed algorithm was statistically compared with other state of the art approaches and was found to outperform other global models, while being statistically similar with local approaches that offer lower interpretation capabilities.

Estimating forest soil organic carbon content using vis-NIR spectroscopy: Implications for large-scale soil carbon spectroscopic assessment

Large-scale soil organic carbon (SOC) stock assessment is expensive as a large number of samples must be collected and then their time-consuming measurements must be made in the laboratory. Previous studies have shown that visible-near-infrared reflectance (vis-NIR) spectroscopy can quickly predict SOC content at a low cost. However, the application of this method at the large scale remains challenging due to the high spatial heterogeneity of SOC and the spatially dependent relationships of soil spectra and SOC content. Here, we conducted large-scale soil sampling across China's forests and established the Chinese forest soil spectral library (CFSSL) by measuring SOC content and scanning the vis-NIR reflectance of 11, 213 soil samples. Compared with the traditional global partial least squares regression (PLSR) modeling method (R2 = 0.75, RPIQ = 1.95), the clustering by fast research and find of density peak in combination with the Cubist model significantly improved the prediction ability of SOC content (R2 = 0.96, RPIQ = 5.83). This study provided a cost-efficient spectroscopic methodology, including measurement and prediction modeling, for large-scale SOC estimation.

A genetic algorithm‐based stacking algorithm for predicting soil organic matter from vis–NIR spectral data

It has been demonstrated that diffuse reflectance spectroscopy in the visible and near‐infrared (vis–NIR) can be exploited to predict chemical and physical soil properties. Immense soil spectral libraries (SSL) are being developed; therefore, more elaborate tools that capitalize on contemporary knowledge and techniques need to be established to provide accurate predictions. In this paper, we propose a novel genetic algorithm‐based stacking model that makes synergetic use of multiple models developed from different preprocessed spectral sources (termed L1 models). This is a form of ensemble learning where multiple hypotheses are combined to create a more robust and more accurate ensemble hypothesis. The genetic algorithm automatically defines the configuration of the stacked model, by selecting the best cooperating subset of the initial models. Our methodology was tested on the newly developed GEO‐CRADLE SSL to predict soil organic matter (SOM). Results showed that the accuracy of prediction of the proposed method ( R2 = 0.76, and ratio of performance to interquartile range [RPIQ] = 2.22) was better than the one attained by the best L1 model ( R2 = 0.65, RPIQ = 1.93). This approach can thus be effectively utilized to enhance the predictions of soil properties in small and large soil spectral libraries alike.Highlights A novel model stacking algorithm is proposed, combining spectral models from different sources and algorithms. Use of a genetic algorithm is examined, accounting for the large number of possible model permutations. The methodology was applied to the GEO‐CRADLE vis–NIR soil spectral library. Results indicate that model stacking creates more accurate and robust models.

A memory-based learning approach utilizing combined spectral sources and geographical proximity for improved VIS-NIR-SWIR soil properties estimation

The current study was driven by the need to derive improved soil information whereby the strengths of memory-based learning and soil spectroscopy are exploited towards addressing regional challenges and supporting sustainable development across the Balkan, North Africa, and Middle East regions. In this study we focused on a novel derivation of the Spectrum Based Learner (SBL) algorithm by i) taking into account both the geographical proximity and the spectral similarity in the computation of the distance between samples, and ii) using an optimal pair of spectral pre-treatments with the first used for the determination of neighbours while the latter for the estimation of the target property. The proposed approach was tested on a recently developed standardized Soil Spectral Library (SSL) across the VIS-NIR-SWIR spectral region range (350–2500 nm) which comprises 1760 soil samples from 9 different countries and is considered the largest and most diverse soil database infrastructure in the region compatible with other SSLs. Here, we found that our approach presents a great potential for predicting soil texture contents (Clay, Sand, and Silt), organic carbon (OC), Calcium Carbonate (CaCO3), pH, and Electrical Conductivity (EC). These results outperform the predictive performance of other state of the art global and local approaches that have been applied to similar large and complex soil datasets. The use of geographical coordinates in the computation of the sample similarities enhanced the predictions of soil properties, since it allowed the generation of local subsets that present similar soil compositions. In addition, we conclude that the various spectral sources as derived by a set of predefined spectral pre-processing techniques contain complementary information which should be combined instead of relying solely on the best spectral pre-processing technique. This approach could be effectively utilized to enhance the predictions of soil properties in large and complex SSLs, since it decreased the Root Mean Square Errors (RMSEs) by a relative mean of 6.47% (average value across the properties – decrease ranging from 2.90% to 9.09%) for the various soil properties, compared to other global and local algorithms. We conclude that national SSLs that were measured under a standardization process could further contribute to the global initiative to address challenges and support a data-centric approach for informed decision making with regards to environmental and agricultural issues.

Visisible and near-infrared spectroscopy in Poland: from the beginning to the Polish Soil Spectral Library

Visisible and near-infrared spectroscopy in Poland: from the beginning to the Polish Soil Spectral Library

A Regional Legacy Soil Dataset for Prediction of Sand and Clay Content with Vis-Nir-Swir, in Southern Brazil

ABSTRACT The success of soil prediction by VIS-NIR-SWIR spectroscopy has led to considerable investment in large soil spectral libraries. The aims of this study were 1) to develop a soil VIS-NIR-SWIR spectroscopy approach using legacy soil samples to improve spectral soil information in a regional scale; (2) to compare six spectral preprocessing techniques; and (3) to compare the performance of linear and non-linear multivariate models for prediction of sand and clay content. A total of 1,534 legacy soil samples, stored [...]