Fig 2 - uploaded by Lazarus Chapungu
Content may be subject to copyright.
Examples of mid-infrared spectra of soil samples from the three villages under study: i.e. the baseline-corrected spectrum of one sample per village having an average C content of 7, 11 and 29 g kg − 1 for the communal area, and the old and new resettlement
Source publication
Knowledge of soil spatial variability is important in natural resource management, interpolation and soil sampling design, but requires a considerable amount of geo-referenced data. In this study, mid-infrared spectroscopy in combination with spatial analyses tools is being proposed to facilitate landscape evaluation and monitoring. Mid-infrared sp...
Context in source publication
Context 1
... exchangeable cations and effective CEC by extraction with ammonium chloride (Schöning and Brümmer, 2008). All 432 soil samples were analyzed by Diffuse Re fl ectance Infrared Fourier Transform (DRIFT)-MIRS. Five grams of ball-milled soil samples were scanned in a TENSOR-27 FT-IR spectrometer (Bruker Optik GmbH, Germany) coupled to a DRIFT-Praying Mantis chamber (Harrick Scienti fi c Products Inc., New York, US). Spectra were obtained at least in triplicate, from 600 to 4000 wavenumber cm − 1 , with a resolution of 4 cm − 1 and 16 scans/sample, and expressed in absorbance units [log(1 / Re fl ectance)]. Potassium bromide (KBr) for IR spectroscopy (assay ≥ 99.5%), kept always dry in a desiccator, was used as a background. All spectral replicates per sample were averaged and later subjected to multivariate calibration by using partial least square (PLS) regression, which relates the processed spectra (e.g. Fig. 2) to the related concentration values from the reference samples. Through a random split selection of the reference samples, half of the samples were used for calibration, while the other half left for validation. Chemometric models were constructed with the “ optimization ” function of the OPUS-QUANT2 package (Bruker Optik GmbH, Germany). Calibration regions were set to exclude the background CO 2 region (2300 – 2400 cm − 1 ) and the edge of the detection limits of the spectrometer ( b 700 and N 3900 cm − 1 ) to reduce noise. Prediction accuracy of selected MIRS models was evaluated by the residual prediction deviation (RPD) value, the coef fi cient of determination ( R 2 ) and the root mean square error of the prediction (RMSEP). Once suitable chemometric models were selected, models were applied to every spectrum replicate of non- reference samples for the prediction of unknown concentration values for each possible soil property; and results of all replicates per sample were fi nally averaged. All spectral manipulation and development of chemometric models were carried out in OPUS, version 6.5 (Bruker Optik GmbH, Germany). Descriptive statistics were calculated to explore the distribution of each soil property under evaluation and as a critical step before geostatistical analyses (Olea, 2006). This comprised the calculation of univariate statistical moments (e.g. mean, median, range), construction of scatter plots, box plots, frequency tables and normality tests, as well as the identi fi cation of true outliers and their exclusion if necessary, as even a few outliers can produce very unstable results (Makkawi, 2004). We usually considered as outliers those points with values higher or lower than three standard deviations from the mean (Liu et al., 2009). The coef fi cient of variation (CV) was calculated as an index for assessing overall variability (Gallardo and Paramá, 2007). The non-parametric tests of Kruskal – Wallis and Mann – Whitney (a Kruskall – Wallis version for only two levels) were chosen for testing the equality of medians among villages following the method of Bekele and Hudnall (2006). All classical statistical analyses were performed in SAS version 9.2 (SAS Institute Inc). The minimum number of samples required for estimating the mean of the different evaluated soil properties in each village, at different probabilities of its true value (error), with a 95% of con fi dence, was estimated by using Eq. ...
Citations
... Some soil properties, such as texture, pH, and porosity, are considered to be rather spatially static, while other features such as soil nitrogen N, soil available forms of P and K, and biological properties are highly spatially variable (Piotrowska & Długosz 2012). Soil spatial variability can occur across multiple spatial scales, ranging from the micro level (millimeters) to the plot level (meters) and up to the landscape level (kilometers) (Cobo et al., 2010). The aim of the study was to determine the physical and chemical properties of soils found on Toposequence along University of Benin. ...
The experiment was conducted at University of Benin, Nigeria, involving soil samples from four toposequence sites (Crest, Middle, Lower, and Bottom) at different depths (0-15 cm, 15-30 cm, and 30-45 cm). A total of 36 samples were collected and analyzed for various parameters using standard procedures. The parameters included particle size distribution, pH, total organic carbon (TOC), total nitrogen (N), available phosphorus (P), Carbon (C), Hydrogen (H), Magnesium (Mg), Potassium (K), Sodium (Na), ECEC, and Aluminum (Al). Results indicated that pH was lowest in the Crest area (pH 4.10 at 30-45 cm depth) and highest in the Bottom area (pH 5.80 at 0-15 cm to 30-45 cm depth). Different soil properties showed varying highest values across the toposequence depths. These properties included Total organic C, Total N, Available P, Ca, K, Mg, H, Na, ECEC, sand content, and the various forms of phosphorus. The correlation table revealed significant positive and negative relationships between different forms of phosphorus and various soil physical and chemical properties. The experiment demonstrated distinct variations in soil properties along the toposequence sites and depths. The findings contribute to a better understanding of soil characteristics in the studied region, aiding in informed agricultural practices and land management decisions.
... In addition, the TP cannot be predicted quantitatively mainly because there was no direct feature band on both spectra regarding the TP, while the correlation between the TP and TN was not significant. Similar results were reported by some previous studies [60][61][62]. ...
As a precious soil resource, black soils in Northeast China are currently facing severe land degradation. Visible and near-infrared spectroscopy (vis-NIR, 350–2500 nm) and mid-infrared spectroscopy (MIR, 2500–25,000 nm) have shown great potential to predict soil properties. However, there is still limited research on using MIR in situ. The aim of this study was to explore the feasibility of in situ MIR for the prediction of soil total nitrogen (TN) and total phosphorus (TP) and to compare its performance with the use of laboratory MIR, as well as the use of in situ and laboratory vis-NIR. A total of 450 samples from 90 soil profiles, along with their in situ and laboratory spectra of MIR and vis-NIR, were collected in a field with ten different tillage and management practices in a typical black soil area of Northeast China. Partial least square regression (PLSR), random forest (RF) and multivariate adaptive regression splines (MARS) were used to generate the calibrations between the spectra and the two properties. The results showed that both MIR and vis-NIR were able to predict the TN whether in laboratory or in situ conditions, but neither of them could predict the TP quantitatively since there was no sensitive band on both spectra regarding the TP. The prediction accuracy of the TN with laboratory spectra was higher than that with in situ spectra, for both vis-NIR and MIR. The optimal prediction accuracy of the TN with laboratory MIR (RMSE = 0.11 g/kg, RPD = 3.12) was higher than that of laboratory vis-NIR (RMSE = 0.14 g/kg, RPD = 2.45). The optimal prediction accuracy of in situ MIR (RMSE = 0.20 g/kg, RPD = 1.80) was lower than that of in situ vis-NIR (RMSE = 0.16 g/kg, RPD = 2.14). The prediction performance of the spectra followed laboratory MIR > laboratory vis-NIR > in situ vis-NIR > in situ MIR. The performance of in situ MIR was relatively poor, mainly due to the fact that MIR was more influenced by soil moisture. This study verified the feasibility of in situ MIR for soil property prediction and provided an approach for obtaining rapid soil information and a reference for soil research and management in black soil areas.
... For example, Guo et al. [18] found that the most common characteristic wavelengths for AK in paddy soil were located around 400-483 nm, 728 nm, 967-1031 nm, 1271-1409 nm, 1643-1789 nm, 1975-2004 nm, 2109-2174 nm, and 2312-2449 nm. The differences in soil AK characteristic wavelengths across different regions can be a ributed to factors such as soil heterogeneity, moisture content, texture, color, sample number, sample and spectral data pretreatment, concentration range, and the model development method [17,25,69]. ...
... Although it could represent the characteristics of local farmland, based on soil sampling data, it was not adequate to build a universal inversion model of soil AK content. In addition, environmental factors are commonly crucial to the inversion of soil AK content [21,69], and the relationships among environmental factors, spectral data and inversion models were not thoroughly discussed in this paper. ...
Estimating the available potassium (AK) in soil can help improve field management and crop production. Fourier-transform infrared (FTIR) spectroscopy is one of the most promising techniques for the fast and real-time analysis of soil AK content. However, the successful estimation of soil AK content by FTIR depends on the proper selection of appropriate spectral dimensionality reduction techniques. To magnify the subtle spectral signals concerning AK content and improve the understanding of the characteristic FTIR wavelengths of AK content, a total of 145 soil samples were collected in an agricultural site located in the southwest part of Sichuan, China, and three typical spectral dimensionality reduction methods—the successive projections algorithm (SPA), simulated annealing algorithm (SA) and competitive adaptive reweighted sampling (CARS)—were adopted to select the appropriate spectral variable. Then, partial least squares regression (PLSR) was utilized to establish AK inversion models by incorporating the optimal set of spectral variables extracted by different dimensionality reduction algorithms. The accuracy of each inversion model was tested based on the coefficient of determination (R2), root mean square error (RMSE) and mean absolute value error (MAE), and the contribution of the inversion model variables was explored. The results show that: (1) The application of spectral dimensionality reduction is a useful technique for isolating specific components of multicomponent spectra, and as such is a powerful tool to improve and expand the predicted potential of the spectroscopy of soil AK content. Compared with the SA and CARS algorithms, the SPA was more suitable for soil AK content inversion. (2) The inversion model results showed that the characteristic wavelengths were mainly around 777 nm, 1315 nm, 1375 nm, 1635 nm, 1730 nm and 3568–3990 nm. (3) Comparing the performances of different inversion models, the SPA–PLSR model (R2= 0.49, RMSE = 22.80, MAE = 16.82) was superior to the SA–PLSR and CARS–PLSR models, which has certain guiding significance for the rapid detection of soil AK content.
... At the same time, climate and land use management facilitate the erosional processes influenced by relief and soil properties. It is, however, shared knowledge that anthropogenic activities have amplified the rate of landscape modification through soil erosional processes (Baade and Glotzbach, 2017;Cobo et al., 2010). Humans continue to modify natural conditions through management practices. ...
Soil erosion has become a common feature in certain national parks across the global biomes, and national parks in the Savanna biome of South Africa have not been spared. Given their prominence in hosting large land mammals that contribute significantly to the tourism sector, the vulnerability of the parks to soil erosion merits attention. This study examines soil erosion status in the Savanna biome national parks of South Africa to recommend management practices aligned with the degree of erosion susceptibility. The results show that erosion rates across the national parks range between 0 and 25 t/ha/yr, and only a few have soil loss rates above 25 t/ha/yr. The Kgalagadi National Park has the lowest erosion rates, with a water erosion severity index (ESI) of 0. The Kruger and Mokala National Parks have low erosion rates, with an ESI of 0.017 and 0.01, respectively. The Mapungubwe National Park has moderate rates of soil loss, with an ESI of 0.023, while Marakele has the highest rates of soil erosion, with an ESI of 0.39. Soil erosion by water action is generally a minor environmental threat for the Kgalagadi, Kruger and Mokala National Parks. However, for Mapungubwe, soil erosion has a moderate threat, while in Marakele, it imposes a very high threat level. The different rates of soil erosion are attributed to the heterogeneous biophysical conditions of the Savanna biome, as well as different historical land use patterns across the parks. Climate change could be exacerbating the susceptibility of national parks to soil erosion. Park-specific management strategies that consider historical land use patterns and biophysical conditions within a climate change context are recommended.
... We collected soil samples from the study area in October-November 2020. Terrain, vegetation, soil, land-use type, and road accessibility were taken into account to ensure the representativeness and spatial independence of the samples (Cobo et al., 2010). Four to six sampling points were selected in each county-level administrative region, with a total of 512 sampling points (Figure 1). ...
Visible–near‐infrared (Vis‐NIR) spectroscopy is increasingly used to predict soil organic carbon (SOC) content. However, the prediction accuracy of this technology is dependent on model selection and study scale. This study explored the roles of spectral variable selection and stratified calibration based on soil type in Vis‐NIR spectroscopy for predicting SOC content at a provincial scale. A total of 490 samples, collected in Jiangxi Province (southeast China), were used for modeling with partial least squares regression, support vector machine, random forests, and back‐propagation neural network (BPNN). The feature wavebands of soil samples were selected by competitive adaptive reweighted sampling (CARS), and a stratified calibration was conducted based on soil type. The results showed that CARS‐based models outperformed models with full wavebands in predicting the SOC content. The CARS‐BPNN model combined with stratified calibration showed the best prediction performance for total soils (validation set R² = .82, which was .21 higher than that of BPNN based on global calibration). This study established an accurate method to predict SOC content from provincial‐scale spectral data using the CARS‐BPNN model coupled with stratified calibration based on soil type.
... The coefficient of determination (R 2 ), Eq. (1), residual prediction deviation (RPD), Eq. (2), root mean square error (RMSE), Eq. (3), were used to evaluate the prediction ability of the MIR spectroscopy technique are presented in Table 2. The models of calibration and validation showed the prediction potential of MIR [8]. ...
... This implies validation of the model in terms of P and S determination was poor. Similar results were reported by [11,21,22,23], and [8] reported similar findings. This insufficient validation for P and S is due to their low concentration in the soil. ...
Knowledge of soil nutrient status is a basic requirement in sustainable agriculture. However, assessment of soil properties has long been done through conventional laboratory analysis, which is costly and time-consuming. Therefore, developing alternative, cheaper and faster techniques for soil analysis is highly required. Mid-infrared spectroscopy (MIRS) techniques are rapid, convenient, environmentally friendly, and nondestructive techniques for quantifying several soil properties. This study aimed to evaluate the prediction performance of MIR for pH, organic carbon (O.C.), available phosphorus and sulfur, total nitrogen, exchangeable cations, and micronutrient. Soil samples were collected from southern Ethiopia. In this study, properties of 3882 soil samples were used as references from different parts of Ethiopia. Partial least squares regression (PLSR) was used for calibration. The correlation of measured and predicted properties of soil samples collected was analyzed using the Pearson correlation coefficient. Better prediction was obtained for Ca (R²=0.95 and RPD=3.9), CEC (R²=0.92 and RPD=3.5), TN (R²=0.92 and RPD=3.4), OC (R²=0.91 and RPD=3.4), Mg (R²=0.84 and RPD=2.6), pH (R²=0.85 and RPD=2.4) and Fe (R²=0.65 and RPD=1.7). In general, soil properties could be predicted using MIRS methods. On the other hand, soil nutrients that showed poor prediction require further studies.
... MIRS, in combination with Partial least square regression (PLSR), has proven its potential for the rapid evaluation of various soil chemical Table 2 Absorption bands and functional group predicted on literature research ( Matamala et al., 2019 ). ( Nguyen et al., 1991 ;Nayak and Singh, 2007 ;Rossel and Behrens, 2010 ;Churchman et al., 2010 ;Veum et al., 2014 ;Artz et al., 2008 ;Soriano-Disla et al., 2014 ;Ellerbrock and Gerke, 2004 ;Haberhauer and Gerzabek, 1999 properties such as SOC, soil pH, N, soil texture, and soil carbon fractions ( Janik and Skjemstad, 1995 ;Nath et al., 2021 ;Bornemann et al., 2010 ;Cobo et al., 2010 ;D'Acqui et al., 2010 ;Zimmermann et al., 2007 ;Bornemann et al., 2008 ;Du et al., 2008b;Ludwig et al., 2008;Sudduth et al., 2009;Sudduth and Hummel, 1993 ). The total amount of TOC and TN in the soil were influenced by temperature. ...
Evaluating the all-encompassing idea of soil quality requires an enormous set-up of estimations of soil physical, chemical, and biological properties. The fast and appropriate evaluation of soil quality (SQ) is expected to keep up soil wellbeing. Appraisal of soil quality is currently turning into a standard work for crop production and management of soil. Nevertheless, regular research centre investigation dependent on the assurance of soil properties, which is time and cost devouring and is not reasonable for precision agriculture. Mid-infrared spectroscopy is a quick, cheap, ecological agreeable, non-dangerous, and repeatable method. We observed a fantastic presentation of foreseeing soil C and N substance utilizing mid-infrared spectra. In a large portion of cases, the expectations of soil N, P, K, S, and few micronutrients are acceptable. Clay minerals, soil water (volumetric and gravimetric water content), and soil microorganisms can likewise be demonstrated and assessed utilizing mid-infrared spectroscopy. "Here, we review the possibility of mid-infrared spectroscopy to assess soil quality. The significant focuses are the accompanying: 1) predict physical properties of soil 2) predict chemical properties of soil 3) predict biological properties of soil with mid-infrared spectroscopy". Physical properties of soil that depend on void connection, for example, hydraulic conductivity (HC), bulk density (BD), as well as water-holding capacity (WHC) and can't be anticipated well utilizing mid-infrared spectra. Whereas properties dependent on the soil composition, for example, percentage of clay and shrinking- swelling properties, can be expected sensibly well. Even though presently using mid-infrared spectroscopy in the estimation of soil properties is constrained, it seems favourable to gauge the quality of soil reliant on spectral pre-treatment and multivariate alignment because of the powerful intervene in the spectral band. Partial least square (PLS) is a broadly applied statistical tool to enhance various soil properties' prediction operations. A spectral library and a standard protocol can be an option in the future so that proper utilization of the mid-infrared spectral region is possible. Accordingly, mid-infrared spectroscopy coupled with a suitable statistical tool can assess soil quality, which will play a significant role in maintaining agriculture sustainability.
... Therefore, the soil environmental parameter data obtained by spatial interpolation are less reliable for soil salinity, but they have a higher reference value for soil texture. The sampling scale affects the spatial variability of soil properties, and the degree of autocorrelation of variables in the same landscape can vary considerably at different scales [74,75]. As the scale increases, the forms of variation in soil properties tend to become more complex, and it is possible that their homogeneity at a small scale is an important cause of their variability at large scale [76,77]. ...
The purpose of this paper was to study the spatial characteristics and possible influencing factors of farmland soil texture and salt content in the Syr Darya River Basin. Data on the soil grain size and salt content were collected at 56 sampling sites in the southern part of the Shardara Reservoir and the left bank of the Syr Darya River irrigation area. With the methods of local spatial statistics (Getis-Ord Gi* and Moran's I), the hotspots of soil salinity and grain size in the study area were revealed, and along with the use of correlation analysis, the possible factors affecting soil salt distribution were discussed. Among the 56 soil sampling sites, sandy loam, loamy loam, and chalky loam accounted for 20%, 50%, and 30%, respectively, and mildly, moderately, and severely saline soils accounted for 80.36%, 14.28%, and 5.36%, respectively. There was statistically significant spatial autocorrelation between sand, silt, and clay content in the soils, but the spatial autocorrelation for salt content was weak. The results show that high and high-cluster areas (hotspots) with statistically significant salt content are mainly distributed in the northwest of the study area and that the hotspot distribution of salt content is mainly affected by topography (altitude), but the effect of soil texture on salt content is not significant. The control of soil salinity should prioritize low-altitude areas, especially in the northwestern region. The results are of great significance for the regulation and control of soil salinity and the sustainable utilization of soil in arid Central Asia.
... A possible alternative is the use of techniques such as soil spectroscopy to rapidly assess soil weathering degree for a large number of samples at reduced cost. Soil spectroscopy has been used to estimate numerous soil properties such as organic matter, oxides, texture or exchangeable cations (Calderón et al., 2011;Cobo et al., 2010;Janik et al., 1998;Madari et al., 2006;Richter et al., 2009). However, few studies have investigated the application of soil spectroscopy for the determination of weathering indices (Baptista et al., 2011;Mohanty et al., 2016;Zhao et al., 2018). ...
... For the other major oxides considered in the CIA calculation both poorer predictions (Janik et al., 1995) and predictions of similar quality (Janik et al., 1998) were reported. Several studies already showed the potential of MIR reflectance analysis to predict exchangeable cations, including exchangeable Ca, Mg, K and Na (Cobo et al., 2010;Janik et al., 1995;Minasny et al., 2009;Pirie et al., 2005;Soriano-Disla et al., 2014). Exchangeable Na and K sometimes showed lower correlations, probably due to narrow ranges and low concentrations (Janik et al., 1998;Minasny et al., 2009;Pirie et al., 2005). ...
Land use and land cover changes (LUCC) can drastically alter various components of the critical zone, including soil thickness and soil chemical weathering processes. Often these studies, however, tend to focus on extreme cases, not representing what actually happens on average at larger, regional scales. Here, we evaluate the impact of LUCC on soil thickness and soil weathering degree at the regional scale, where we use soil spectroscopy to derive weathering indices. In a subtropical region in Southern Brazil, we collected calibration/validation soil samples (n = 49) from 4 different locations for which we measured the mid-infrared (MIR) spectral reflectance and 3 soil chemical weathering indices: chemical index of alteration (CIA), the total reserve in bases (TRB), and the iron ratio (Fed/Fet). We used partial least square regressions on this calibration/validation dataset to relate the MIR spectra of the soil samples to these weathering indices, resulting in good calibration relationships with R² values of 0.97, 0.91 and 0.84 for CIA, TRB and Fed/Fet, respectively. Applying these relations to MIR spectra of regionally collected soil samples allowed us to calculate soil weathering degrees for a large number of soil samples (n = 229), without requiring costly and time-consuming chemical analyses. We collected these soil samples at 100 mid-slope positions: 50 under forest and 50 under agricultural land use. Land use explained only a minor part of the variation in soil thickness and weathering degree. Thus, while local water and tillage erosion rates might be considerable after deforestation, this has not led to significant reductions in average soil thickness and has not affected soil weathering degree. Slope gradient is the main factor influencing the spatial variability in soil thickness and weathering degree on mid-slope sections in our study area. Human activities over the last century did not fundamentally alter these patterns.
... Specifically, in this paper, we collect spectral data while discounting for thermal heating of the sample, perform a spatial variability analysis, and model the probability distribution of this data as a Bayesian hierarchical model. This approach is prevalent in geostatistics and has also been used before with MIR data for soil composition assessment [11]. However, to the best of our knowledge, this approach has never been tested on smaller-scale experiments. ...
The problem of analyzing substances using low-cost sensors with a low signal-to-noise ratio (SNR) remains challenging. Using accurate models for the spectral data is paramount for the success of any classification task. We demonstrate that the thermal compensation of sample heating and spatial variability analysis yield lower modeling errors than non-spatial modeling. Then, we obtain the inference of the spectral data probability density functions using the integrated nested Laplace approximation (INLA) on a Bayesian hierarchical model. To achieve this goal, we use the fast and user-friendly R-INLA package in $ R $ R for the computation. This approach allows affordable and real-time substance identification with fewer SNR sensor measurements, thereby potentially increasing throughput and lowering costs.
