No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Near infrared reflectance spectroscopy for assessment of spatial soil variation in an agricultural field
Abstract
Efficient tools to measure within-field spatial variation in soil are important when establishing agricultural field trials and in precision farming. The object of the study was to investigate if a combination of two techniques, principal component analysis (PCA) and geostatistics, could reveal spatial soil variation from near infrared reflectance (NIR) spectroscopy data and thereby replace more conventional, viz. laborious and expensive, soil analyses. NIR spectrum is known to reveal information about important soil chemical, physical and biological properties and has been used in soil science for years. Three soil variables, total carbon (Tot-C), clay content and pH, were used as reference variables. The study was carried out on one site (200×160 m) in eastern Sweden with a Eutric Cambisol soil type where a sampling grid of 20×20 m was established. From the grid nodes, 99 samples were collected to a depth of 10 cm. The soil was analyzed by NIR and the data were decomposed by PCA. The first two principal components (PC 1 and PC 2) explained 85% of the total variance and therefore these two PCs were selected for further assessment of spatial variation by variography and kriging. PC 1 showed the strongest spatial dependence with a range of 148 m and a nugget close to zero. The variogram for PC 1 was robust and the kriging map expressed a clear pattern. The range of spatial correlation varied between the three reference soil variables. Tot-C expressed a low spatial dependence with a high proportion of nugget, whereas clay content and pH expressed spatial dependence at a range of 54 and 46 m, respectively. Neither of the traditional soil variables showed as strong spatial dependence as PC 1 of NIR. The advantage of the NIR–PCA strategy is that the first PCs will capture the spectral bands that express the largest variation regardless of what the NIR bands correlate to and, hence, PC 1 will always explain the variation of the soil properties that in each specific case have the largest influence on the PCA model. In conclusion, the NIR–PCA strategy seems to be an efficient and reliable strategy to use when determining the soil spatial variation in a field.
... Spektral data almadan önce, bazen alınacak verinin kalitesini arttırmak amacı ile örneklerin fiziksel olarak (ufalama gibi) ön muameleye tabi tutulmaları gerekebilir. Günümüze kadar yapılmış olan bir çok çalışmada görülebilir, yakın ve orta kızıl ötesi bölgedeki ışıma; organik madde (Ben-Dor ve Banin, 1995; Shepherd ve Walsh, 2004; Islam ve ark., 2005;Brown ve ark., 2006; McBratney ve ark., 2006), toplam azot (Madari ve ark., 2006), toplam karbon (Odlare ve ark., 2005; Madari ve ark., 2006; Reeves III ve ark., 2006), yarayışlı fosfor (Bogrekci ve Lee, 2005; Mouazen ve ark., 2007), demir oksitler (Ben- Dor ve Banin, 1995; Dematte ve ark., 2004; Brown et al., 2006); kil mineralleri (Dematte ve ark., 2004; Brown ve ark., 2006);toprak parçacık büyüklük dağılımı (Cozzolino ve Moron, 2003; Islam ve ark., 2003 ve 2005; Odlare ve ark., 2005; Sorensen ve Dalsgaard, 2005; Mouazen ve ark., 2005; Madari ve ark., 2006; Brown et al., 2006; McBratney ve ark., 2006), kil mineral tipi (Brown ve ark., 2006), katyon değişim kapasitesi (Brown ve ark., 2006; McBratney ve ark., 2006), pH (Islam ve ark., 2005; Odlare ve ark., 2005; McBratney ve ark., 2006), nem içeriği ve agregatlaşma (Hummel ve ark., 2001; Mouzen ve ark., 2006a) gibi toprak özeliklerinin belirlenmesinde, bu özelliklerin ışımaya olan doğrudan etkilerinden dolayı başarılı bir şekilde kullanılabileceği gösterilmiştir. Viscarra Rossel ve ark. ...
... Spektral data almadan önce, bazen alınacak verinin kalitesini arttırmak amacı ile örneklerin fiziksel olarak (ufalama gibi) ön muameleye tabi tutulmaları gerekebilir. Günümüze kadar yapılmış olan bir çok çalışmada görülebilir, yakın ve orta kızıl ötesi bölgedeki ışıma; organik madde (Ben-Dor ve Banin, 1995; Shepherd ve Walsh, 2004; Islam ve ark., 2005;Brown ve ark., 2006; McBratney ve ark., 2006), toplam azot (Madari ve ark., 2006), toplam karbon (Odlare ve ark., 2005; Madari ve ark., 2006; Reeves III ve ark., 2006), yarayışlı fosfor (Bogrekci ve Lee, 2005; Mouazen ve ark., 2007), demir oksitler (Ben- Dor ve Banin, 1995; Dematte ve ark., 2004; Brown et al., 2006); kil mineralleri (Dematte ve ark., 2004; Brown ve ark., 2006);toprak parçacık büyüklük dağılımı (Cozzolino ve Moron, 2003; Islam ve ark., 2003 ve 2005; Odlare ve ark., 2005; Sorensen ve Dalsgaard, 2005; Mouazen ve ark., 2005; Madari ve ark., 2006; Brown et al., 2006; McBratney ve ark., 2006), kil mineral tipi (Brown ve ark., 2006), katyon değişim kapasitesi (Brown ve ark., 2006; McBratney ve ark., 2006), pH (Islam ve ark., 2005; Odlare ve ark., 2005; McBratney ve ark., 2006), nem içeriği ve agregatlaşma (Hummel ve ark., 2001; Mouzen ve ark., 2006a) gibi toprak özeliklerinin belirlenmesinde, bu özelliklerin ışımaya olan doğrudan etkilerinden dolayı başarılı bir şekilde kullanılabileceği gösterilmiştir. Viscarra Rossel ve ark. ...
... Spektral data almadan önce, bazen alınacak verinin kalitesini arttırmak amacı ile örneklerin fiziksel olarak (ufalama gibi) ön muameleye tabi tutulmaları gerekebilir. Günümüze kadar yapılmış olan bir çok çalışmada görülebilir, yakın ve orta kızıl ötesi bölgedeki ışıma; organik madde (Ben-Dor ve Banin, 1995; Shepherd ve Walsh, 2004; Islam ve ark., 2005;Brown ve ark., 2006; McBratney ve ark., 2006), toplam azot (Madari ve ark., 2006), toplam karbon (Odlare ve ark., 2005; Madari ve ark., 2006; Reeves III ve ark., 2006), yarayışlı fosfor (Bogrekci ve Lee, 2005; Mouazen ve ark., 2007), demir oksitler (Ben- Dor ve Banin, 1995; Dematte ve ark., 2004; Brown et al., 2006); kil mineralleri (Dematte ve ark., 2004; Brown ve ark., 2006);toprak parçacık büyüklük dağılımı (Cozzolino ve Moron, 2003; Islam ve ark., 2003 ve 2005; Odlare ve ark., 2005; Sorensen ve Dalsgaard, 2005; Mouazen ve ark., 2005; Madari ve ark., 2006; Brown et al., 2006; McBratney ve ark., 2006), kil mineral tipi (Brown ve ark., 2006), katyon değişim kapasitesi (Brown ve ark., 2006; McBratney ve ark., 2006), pH (Islam ve ark., 2005; Odlare ve ark., 2005; McBratney ve ark., 2006), nem içeriği ve agregatlaşma (Hummel ve ark., 2001; Mouzen ve ark., 2006a) gibi toprak özeliklerinin belirlenmesinde, bu özelliklerin ışımaya olan doğrudan etkilerinden dolayı başarılı bir şekilde kullanılabileceği gösterilmiştir. Viscarra Rossel ve ark. ...
ZET:Hassas tarım tekniklerinin uygulanması, küresel olarak toprakta karbon zenginleşmesinin gözlemlenmesi ve toprak kalitesinin sürdürülebilirliğini sağlayacak toprak özelliklerinin daha hızlı belirlenebileceği, ucuz ve güvenilir yöntemlere olan gereksinim sürekli artmaktadır. Toprakların fiziksel, kimyasal, biyolojik ve mineralojik özelliklerinin mevcut laboratuar yöntemler ile belirlenmesi pahalı ve oldukça zaman ve işçilik gerektirdiği gibi, analiz için kullanılan güçlü kimyasalların atıkları çevreye zarar verebilmektedir. Geleneksel olarak kullanılan laboratuar yöntemlerine alternatif olarak son zamanlarda yaygın bir şekilde kullanılmaya başlanan dağılmış yansıma spektroskopi (morötesi, görülebilir, yakın kızıl ötesi ve orta kızıl ötesi) tekniği, pH, organik karbon, su içeriği, parçacık büyüklük dağılımı, katyon değişim kapasitesi, değişebilir katyonlar, kil mineralojisi ve daha bir çok toprak özelliğinin hızlı bir şekilde belirlenmesine olanak vermektedir. Yansıma özelliklerinden gidilerek toprak özelliklerinin belirlenmesinde gelişmiş istatistiksel yöntemlerden faydalanılmaktadır. Çoklu regresyon analizi, temel bileşenler analizi, kısmi en az-karelerin regresyonu ve sinir ağları kalibrasyonu yaygın olarak kullanılan yöntemlerdir. Toprak özelliklerinin bozulmadan, yerinde incelenebilmesine olanak veren taşınabilir spektroskopi cihazları, arazideki değişkenliğin daha güvenilir şekilde incelenmesine olanak tanımaktadır. Tüm bu avantajlarının yanında, spektroskopik yöntemlerin doğruluğu kalibrasyona ve kullanılan referans metodun hassasiyeti ve doğruluğuna oldukça bağlıdır. Bu nedenle aletin kalibrasyonunda doğruluğu kabul edilmiş olan referans metotların kullanılması kaçınılmazdır. ABSTRACT:Application of precision agriculture techniques, monitoring of carbon sequestration in soils throughout the world and sustaining the soil quality require reliable, fast and cheap soil analysis techniques. The determination of soil physical, chemical, biological and mineralogical characteristics with conventional laboratory analysis can be great time and labor consuming and expensive. The waste of strong chemicals used in soil analysis is hazardous to environment. Diffuse reflectance spectroscopy (ultraviolet, visible, near infrared, mid infrared) as an alternative to conventional laboratory methods has been recently used to determine soil characteristics (soil pH, organic carbon, water content, particle size distribution, cation exchange capacity, exchangeable cations, clay mineralogy and many others) rapidly and inexpensively. Spectroscopic methods require the development of calibrations that relate the spectral information to the property of interest using several statistical methods. Multiple regression analysis, principal component analysis, partial least square regression and neural network are the commonly used multivariate statistical procedures. Portable spectroscopy equipments allow in situ characterization of soil characteristics; thereby variability of soil properties can be also determined in the field. Furthermore, the accuracy of spectroscopic techniques depends on the calibration and the precision and accuracy of the reference method. Therefore, reliable analytical methods need to be used in calibration of spectroscopic technique used in the analysis.
... In NIR domain (12500 to 4000 cm -1 ), overtones and combinations of the fundamental vibrations due to the stretching and bending of OH groups dominate. Since two decades, NIR spectroscopy has being developed for soil property analysis like clay and organic matter content, organic carbon, pH and cation exchange capacity (Ben-Dor et al., 1997;Viscarra Rossel and McBratney, 1998;Odlare et al., 2005;McBratney et al., 2006;Brown, 2007;Waiser et al., 2007). Thus, NIR spectroscopy appears as a major tool in soil sciences for precision agriculture Odlare et al., 2005;Waiser et al. 2007) and ...
... Since two decades, NIR spectroscopy has being developed for soil property analysis like clay and organic matter content, organic carbon, pH and cation exchange capacity (Ben-Dor et al., 1997;Viscarra Rossel and McBratney, 1998;Odlare et al., 2005;McBratney et al., 2006;Brown, 2007;Waiser et al., 2007). Thus, NIR spectroscopy appears as a major tool in soil sciences for precision agriculture Odlare et al., 2005;Waiser et al. 2007) and ...
... The relation between the integrated intensity of absorption features in the characteristic wavelength and the concentration of the chemical group responsible for the absorption is used.For preventing swelling soil damage, estimations of clay minerals relative abundances at large scales allow to identify the high-risk areas to focus geotechnical analyses. Some studies have already showed that it is possible to have information about the clay content of soil(Ben-Dor and Banin 1995;Odlare 2005; ...
The purpose of this study is the detection and quantification of clay minerals, especially swelling clay minerals, in surface soils part. This study is carried out at local and global scale, in order to support swelling soil hazards approach of the engineering offices. The first part presents general principles of infrared spectroscopy and the database used in the thesis. Then, we compare two frequently used methods in soil sciences: statistic methods (PLSR and PCA) and absorption figures analysis featuring clay minerals. They mainly provide kaolinite quantification in multiphase mixtures and in soil. The limits of smectite quantification of both methods lead to a new method development based on continuous wavelet transform and cross correlation. This method is the most efficient and promising one for weak quantification of swelling minerals, at local scale as well as at global scale, in studies prevention of swelling soil hazard
... In NIR domain (12500 to 4000 cm -1 ), overtones and combinations of the fundamental vibrations due to the stretching and bending of OH groups dominate. Since two decades, NIR spectroscopy has being developed for soil property analysis like clay and organic matter content, organic carbon, pH and cation exchange capacity (Ben-Dor et al., 1997;Viscarra Rossel and McBratney, 1998;Odlare et al., 2005;McBratney et al., 2006;Brown, 2007;Waiser et al., 2007). Thus, NIR spectroscopy appears as a major tool in soil sciences for precision agriculture Odlare et al., 2005;Waiser et al. 2007) and ...
... Since two decades, NIR spectroscopy has being developed for soil property analysis like clay and organic matter content, organic carbon, pH and cation exchange capacity (Ben-Dor et al., 1997;Viscarra Rossel and McBratney, 1998;Odlare et al., 2005;McBratney et al., 2006;Brown, 2007;Waiser et al., 2007). Thus, NIR spectroscopy appears as a major tool in soil sciences for precision agriculture Odlare et al., 2005;Waiser et al. 2007) and ...
... The relation between the integrated intensity of absorption features in the characteristic wavelength and the concentration of the chemical group responsible for the absorption is used.For preventing swelling soil damage, estimations of clay minerals relative abundances at large scales allow to identify the high-risk areas to focus geotechnical analyses. Some studies have already showed that it is possible to have information about the clay content of soil(Ben-Dor and Banin 1995;Odlare 2005; ...
... The basic idea of this study is that the variability of the spectra reflects the overall heterogeneity of the biological, chemical and physical properties of the soil and thus can be used as such without preliminary calibration with the variables of interest. In a preceding study, Odlare et al. (2005) found that NIR spectral analysis provides a better description of soil spatial variations than the soil reference variables at the field scale, i.e., in a 3.2-ha experiment plot using the results from principal component analyses realized on the spectral data. Here, we propose to extend this result, and our hypothesis is that raw NIR and MIR spectra can be used as an integrated proxy of many soil properties. ...
... This result is clearly new and generalizes the preceding study proposed by Odlare et al. (2005). In this study, we chose to focus only on chemical properties, even if we acknowledge that a relevant part of soil heterogeneity is driven by physical parameters. ...
Les rémanents de coupe constituent une ressource non exploitée potentiellement utilisable à des fins énergétiques, mais aussi un compartiment essentiel pour la fertilité et la biodiversité des sols forestiers. Des études en zones tropicales ou boréales prouvent les impacts négatifs de telles pratiques sur les écosystèmes forestiers. Le réseau Matières Organiques des Sols (MOS) a été élaboré afin d’évaluer les effets à court et long terme des manipulations de matière organique sur les peuplements forestiers tempérés.Cette thèse a deux objectifs : i) élaborer la méthodologie nécessaire pour caractériser la variabilité des écosystèmes puis mettre en oeuvre le dispositif expérimental national de manipulations de matière organique en tenant compte de cette variabilité et ii) décrire les impacts à très court terme des prélèvements de biomasse sur le fonctionnement des sols forestiers. Ce second objectif s'articule en deux volets: une première partie vise à évaluer l’impact de ces pratiques sylvicoles sur l'équilibre des cycles biogéochimiques et la diversité des communautés fongiques du sol in situ. Une seconde échelle d’étude, en conditions contrôlées, concerne le décryptage des interactions trophiques entre l'arbre, les champignons ectomycorhiziens associés, et les champignons lignivores au cours de la dégradation du bois.Ce premier suivi des sites du réseau MOS nous renseigne sur la réactivité à très court terme des sols forestiers en zone tempérée. L’absence de tendance claire confirme une préservation de l’équilibre des cycles sur un faible laps de temps mais permet de dégager des indicateurs fonctionnels qui semblent répondre rapidement aux perturbations. Un suivi annuel consolidera ces observations sur l’ensemble du dispositif MOS, incluant les indicateurs biologiques qui pourront être mis en évidence sur les communautés fongiques.
... Furthermore, several studies have been focused on developing in situ measurements for soil analysis Christy, 2008;Goldshleger et al., 2010;Morgan et al., 2009;Odlare et al., 2005;Waiser et al., 2007). NIRS is a chemometric-based method that uses all kind of correlations between spectra and direct or indirect chromophores. ...
... Many researchers working on this approach have demonstrated that the optical method is capable of providing low-cost, rapid and reliable products compared to the intensive laboratory work (Arslan et al., 2014;Christy, 2008;Cohen et al., 2005;Dunn et al., 2002;Janik et al., 2009;Reeves III and Smith, 2009;Rossel et al., 2009;Rossel and Webster, 2012;Volkan Bilgili et al., 2010;Zornoza et al., 2008). A number of studies have shown the potential of VNIR-SWIR data analysis using principal component analysis (Odlare et al., 2005;Rizzo et al., 2014;Verheyen et al., 2001) and fuzzy clustering algorithms (Bezdek et al., 1984;De Gruijter et al., 1997;McBratney and de Gruijter, 1992;Odeh et al., 1990;Odgers et al., 2011aOdgers et al., , 2011bVerheyen et al., 2001). Other methods, such as increasing sampling density (Wetterlind et al., 2010) and GIS techniques for modeling soil-generating factors (Karnieli et al., 1998), have proven to be reliable for soil classification based on the relationship between soil properties and soil types. ...
Visible and near-infrared diffuse reflectance spectroscopy is becoming a promising method to estimate various soil properties based on empirical models and spectral libraries. During the past two decades, much attention has been devoted in many scientific fields to capture quantitative information from the reflectance information. In this approach, the reflectance radiation across the Visible and Near Infrared (VNIR) region (400–2500 nm) is modeled against constituents evaluated by wet chemistry methods with the use of various statistical methods. After validation of the optical-chemical model, it can be applied to unknown samples and can provide rapid and objective information about the property in question.
Soil reflectance is easy to use and may provide information about the soil rapidly and simultaneously. Nonetheless, despite its potential, soil spectroscopy has not yet been used to describe soil entity in high-resolution 4D mode (i.e. x, y coordinates, depth, and spectral profile) and its potential has not yet exploited for soil mapping where 4D view is essential.
The purpose of this work is to convert the descriptive information used in the current/traditional soil surveying mission into a quantitative data for soil classification. This goal was achieved by intensive fieldwork of soil sampling, organize hyperspectral aerial campaign, perform laboratory measurements and develop new spectral techniques for analyzing the data.
We have succeeded to evaluate the soil’s “dry spectrum” directly from a wet soil sample in order to predict its clay-content which would be beneficial for future in-situ soil spectral measurements. We managed to perform high resolution 3D soil spectral classification using airborne and field hyperspectral data for soil monitoring. In addition, we were able to assess the spectral detection limit of organic matter in various soil types and also develop cluster-based models for robust assessment of soil properties.
We anticipate that such results will improve the monitoring capacity of different soil characteristics to optimize the spectral analysis for a more accurate soil property prediction and classification.
... However, few infrared spectroscopy studies have attempted to investigate the effect of spatial dependence among soil samples when developing prediction models. Odlare et al. (2005) suggested that combining infrared spectroscopy and geostatistics reveals spatial soil variation and thereby replace the more conventional, laborious and expensive soil analyses. ...
... Geostatistics has been widely used to account for spatial correlation among samples and has been incorporated into soil property prediction models using other auxiliary variables (Odlare et al., 2005). The hybrid method is known as regression-kriging, as distinct from standard regression methods (e.g., PCR, PLS and BRT) or kriging models alone. ...
In this study, the utility of regression-kriging was investigated in building prediction models for soil properties using mid-infrared (7498 to 600 cm− 1) spectral data for soil samples collected from Nyando, Nzoia and Yala catchment areas in Kenya, sampled at 0–20 cm and 20–50 cm depths. Using a systematic technique, 158 samples were selected for analysis of a number of soil properties of interest using wet chemistry methods. We randomly divided the dataset into two groups: 118 samples in the calibration and 40 samples in the holdout validation set. The calibration set was first used to develop partial least squares regression (PLS) models for all the soil properties. Residuals from these models were used to generate semivariograms, which revealed a strong spatial dependence as determined by the ratio of nugget to sill for nitrogen, 9%; Al, 12%; and B, 36%, but with weak spatial dependence for exchangeable Ca (ExCa), 100%; and carbon, 76%. The fitted theoretical semivariograms were used to fit regression-kriging models. Lastly, both the PLS and regression-kriging models were assessed with the validation set and their prediction performance evaluated by R² and root mean square error (RMSE). The results showed that regression kriging method gave lower RMSE values for all the evaluated soil properties except for ExCa, B and exchangeable acidity, with the best predictions, compared with the PLS model, obtained for ExMg (R², 0.93 vs 0.88; RMSE, 6.1 vs 8.4 cmolc kg− 1) and total nitrogen (R² = 0.92 vs R² = 0.74; RMSE, 0.11%, RMSE = 0.2%). In this study, regression-kriging, which takes into account spatial variation normally ignored by other methods, improved use of infrared spectroscopy for predicting soil properties.
... The basic idea of this study is that the variability of the spectra reflects the overall heterogeneity of the biological, chemical, and physical properties of the soil and thus can be used as such without preliminary calibration with the variables of interest. In a preceding study, Odlare et al. (2005) found that NIR spectral analysis provides a better description of soil spatial variations than the soil reference variables at the field scale, i.e., in a 3.2-ha experiment plot using the results from principal component analyses realized on the spectral data. Here, we propose to extend this result, and our hypothesis is that raw NIR and MIR spectra can be used as an integrated proxy of many soil properties. ...
... These results confirm that near-and mid-near-infrared spectra can be used as such in the design of experimental sites as a proxy of a large set of chemical properties of forest soils. This result is clearly new and generalizes the preceding study proposed by Odlare et al. (2005). In this study, we chose to focus only on chemical properties, even if we acknowledge that a relevant part of soil heterogeneity is driven by physical parameters. ...
Key messageNear- and mid-infrared spectroscopy allows for the detection of local patterns of forest soil properties. In combination with dendrometric data, it may be used as a prospective tool for determining soil heterogeneity before setting up long-term forest monitoring experiments. ContextForest soils and stands generally exhibit higher spatial heterogeneity than other terrestrial ecosystems. This variability needs be taken into account before setting up long-term forest monitoring experiments to avoid multiple interactions between local heterogeneity and the factors tested in the experiment. AimsWe hypothesized that raw near- and mid-infrared spectra can be used as an integrated proxy of a large set of soil properties. The use of this method, in combination with dendrometric data, should provide a quick and cost-effective tool for optimizing the design of experimental forest sites. Methods
We assessed the local soil heterogeneity at 11 experimental sites in oak and beech stands, which belong to a new forest long-term ecological research (LTER) network. We used near- and mid-infrared spectroscopy in soil and litter samples. The spectra were subjected to principal components analyses (PCA) to determine the intra-site variability of the soil and litter layers. ResultsBased on mapped PCA coordinates and basic dendrometric data, it was possible to design the experiment and minimize the interactions between the treatment layout and the tested variables. The method was validated with chemical analyses of the soil. No interaction was detected at the set-up of the experiment between the treatment layout and chemical soil properties (C, N, C/N ratio, pH, CEC, Al, Mg, P2O5, Fe, Mn, Na, and K). Conclusion
Near-infrared (NIR) and mid-infrared (MIR) spectroscopy is a useful tool for characterizing the overall heterogeneity of soil chemical properties. It can be used without any preliminary calibration. In combination with dendrometric data, it provides a reliable method for optimizing LTER plots in different types of ecosystems.
... La técnica de espectroscopia de infrarrojo cercano (NIRS) se ha usado para predecir el contenido de Nt, MO, carbono (C) en suelos y su variación espacial (Odlare et al., 2005;Cozzolino y Moron, 2006;Fidencio et al., 2008). Debido a que la técnica NIRS genera numerosos datos de los espectros, es necesario aplicar métodos quimiométricos para interpretar los resultados y generar modelos de predicción para el elemento de interés (Wold et al., 2001). ...
... The technique of near infra-red spectroscopy (NIRS) has been used to predict the content of Nt, organic matter (MO), carbon (C) in soils, and its spatial variation (Odlare et al., 2005;Cozzolino and Moron, 2006;Fidencio et al., 2008). Due to the fact that NIRS technique generates numerous data of spectra, it is necessary to apply chemometric methods in order to interpret the results and generate prediction models for the element of interest (Wold et al., 2001). ...
Nitrogen (N) is one of the most important elements for plant nutrition; therefore, it is necessary to obtain quick and reliable methods for determining N in soil. The objective of the present study was predicting total nitrogen (Nt) concentration in soils of a Mexican tropical region by means of near-infrared spectroscopy (NIRS) in samples within a plastic bag or without one. The advantage of using NIRS lies in the great selectivity of the technique, which makes it possible to quantify a chemical element in a complex mixture without previous work of separation; besides, it allows evaluating Nt amount in few minutes in different soil samples. In the humid tropic of Tabasco, Mexico, 156 soil samples with contrasting chemical characteristics were selected. The samples were dried, ground, and sieved through a 0.5 mm mesh screen, and the Nt concentration was determined by Kjeldahl method. The samples were packed in a plastic bag and Nt was measured in a NIRS spectrophotometer, model FOSS 5000 (NIR systems). The samples were analyzed with and without bag at a level of 1100 to 2000 nm in order to obtain the models which allow assessing Nt amount in soil. The models generated with and without bag were compared in order to recognize whether the interference of the bag can be corrected with the mathematical models and measurements could be made more easily. The models generated for analysis with and without bag explain 92 and 89 % of the variation. Based on this information, it can be concluded that Nt determination with bag is reliable, more rapid, and easier than that in soil without bag, since we can eliminate cleaning the optical fiber of the reading module between one and the next sample.
... The multivariate approach chosen was PCA-analysis. PCA has been widely used in the environmental sciences, for example to investigate drinking water quality (Mouser et al 2005), hazards of landfill leachates (Clement et al, 1997) and spatial variation in the properties of an agricultural field (Odlare and Pell, 2005). ...
... Principal component analysis (PCA) can identify relationships between variables in large datasets by transforming data into new sets of variables, principal components (PC), which allow interpretation of groups of objects with similar characteristics (Odlare, 2005). In the present study, PCA was used to search for relationships between the landfill leachate matrix (i.e. the variables analyzed) and sorption, primarily desorption of phenols and adsorption of heavy metals. ...
In recent years, low cost sorbents, especially pine bark (Pinus Silvestris), have been found effective as adsorbents to remove metals from water. However, major problems have arisen using the material in water treatment facilities for landfill leachate, which have been assumed to depend on the matrix of the landfill leachate. The metal adsorption mechanisms to pine bark were investigated in a multivariate approach using principal component analysis (PCA) and multiple linear regression. Six landfill leachates were treated with pine bark for 24 hours in a batch experiment. The aims were to identify variables that control adsorption, and to investigate whether there are other environmental risks in using pine bark (i.e. metal or phenol desorption from the material). The results showed that there were major problems with treating landfill leachates with low metal concentrations. There was also a significant leaching of Cu and Zn ions from the material. Phenol was also leached from the pine bark. The initial concentration of the metal ion of concern was the most important variable determining the effectiveness of adsorption. Out of the six different landfill leachates treated, only one was effectively cleaned by pine bark. This landfill leachate had a high pH and electric conductivity, and low DOC and high metal concentrations, compared to the other five leachates in the study.
... These results are consistent with previous reports on the usefulness of NIR spectroscopy for predicting soil chemical (Chang and Laird, 2002;Confalonieri et al., 2001) and physical properties (Moron and Cozzolino, 2003). Similarly, Odlarea et al. (2005) found strong correlation between spatial variability in soil properties and NIR spectral data. These same authors report that the joint use of NIR and PC may be efficient tools to measure within-field spatial variation in soil. ...
... Spectral variability can thus be interpreted in terms of PC1 and PC2 alone. Odlarea et al. (2005) previously used NIR spectroscopy to assess spatial variability in agricultural areas and found the variance accounted for by the first two PCs to be 85%. Islam et al. (2005) found the first PCs to explain over 90% of the cumulative variance in most of the studies involving soil spectra reported until then. ...
... The maps of soil contaminated by TPH are produced using high-density soil sampling and traditional laboratory analyses. Nevertheless, such TPH determinations are expensive, extensive laborious, time-consuming, and insufficient when high temporal and spatial resolution of TPH concentrations are guaranteed (Odlare et al., 2005). To analyze and remedy the contamination caused by PHC, there was a lack of rapid and cost-effective assessment methods. ...
Soil petroleum hydrocarbon contamination in the wetlands could cause ecological risk, especially through leakage into water reservoirs. So, the detection of the spatial variability of total petroleum hydrocarbons (TPH) in these soils is very crucial. The variability of TPH and its associations with magnetic susceptibility (χlf) in contaminated soils around the Shadegan pond in southern Iran was investigated. TPH varied from 2.1 to 18.1% (w/w), by the variation of χlf from 14.08 to 713.93 × 10−8 m3 kg−1. The highest variability (coefficient of variation, CV = 107.12%) was obtained for χlf indicating significant impacts of magnetic minerals induced by crude oil contamination. High positive correlations were detected among TPH, χlf, and different forms of iron (Fed: extracted by CBD, Feo: extracted by oxalate, and Fet: total iron). The results of mineralogy by powdery XRD and scanning electron microscopy (SEM), also revealed the formation of ferrimagnetic minerals (magnetite, maghemite) during the biodegradation of petroleum hydrocarbons. The stepwise multiple regression analysis showed that χlf and Fed made a great contribution and could explain about 74% of TPH variability in the studied sites. For the extension of this cost-effective and rapid technique, further work is needed to assay saturation isothermal remnant magnetization and isothermal remanet magnetization in contaminated sites.
... Vis-NIR spectroscopy has also been used to map soils (Bazaglia Filho et al., 2013) and their properties, such as organic matter (Conforti et al., 2013), clay content (Piikki et al., 2013;Ramirez-Lopez et al., 2019), particle size distribution, and cation exchange capacity (Lagacherie et al., 2013;Ramirez-Lopez et al., 2019), as well as to associate several properties with latent variables from the multivariate approach (Webster and Burrough, 1974;Odlare et al., 2005;Rizzo et al., 2016;Chauhan et al., 2021;Sleep et al., 2022). Webster and Burrough (1974), Jang et al. (2020), andChauhan et al. (2021) illustrated the discriminant analysis as one of the best methods to assist in the survey and discretization of soil classes in the field, including the transition limits between them. ...
Visible and near-infrared (Vis-NIR) spectroscopy is a tool to determine soil spatial variability and has been used to map soils and their properties. Considering that physical, chemical, mineralogical, and morphological soil properties can affect the intensity and the depth of the spectral reflectance band in the Vis-NIR region, the objectives of this work were to: (i) evaluate the potential of the reflectance inflection difference (RID) to discriminate soils; and (ii) verify potential spatial correlations of the RID with soil properties, compared with the full spectra, in order to build thematic maps at a field scale. In a farm of 375 ha, 78 soil samples from the 0.87–0.92-m depth were collected in a regular grid of 200 m, with a focus on the soil diagnostic horizon (Bw horizon). The sampled soils were a Latossolo Vermelho-Amarelo ácrico (Haplic Ferralsol) and a Latossolo Vermelho distrófico (Rhodic Ferralsol). Twenty-two physical, chemical, mineralogical, and morphological soil properties were determined, and the Vis-NIR spectra between 400 and 2500 nm were measured. Considering the presence of an inflection band and its relationship with soil properties, the spectral bands used to calculate the RID were between (base 1/base 2): 400–510, 730–930, 1290–1450, 1800–1950, 2000–2218, and 2218–2290 nm. The RID failed to map the spatial variability of soil properties, with a Kappa index of 39%; therefore, it is not a good parameter for building thematic maps of soil parameters. In addition, the complete spectrum (mainly in 400–510, 730–930, 1290–1450, 1800–1950, 2000–2218, and 2218–2290 nm) was better spatially correlated with soil properties than the decomposition of the spectrum by the RID. Soil classification and level of discretization as affected by spectral variability were also discussed here. Three soil groups were discriminated mainly by the Ki and Kr indexes and clay content. Moreover, the variability of the spectra was conditioned by the spatial variability of the mentioned variables. The clay content for soils with a discrepant particle size (group 1 compared with groups 2 and 3) and the Ki and Kr indexes for soils with a homogeneous particle size (between groups 2 and 3), associated with the full Vis-NIR spectral analysis, allowed building thematic maps with a good precision, without the need of mathematical models; this was possible by the modification of the reflectance intensity and the size of the concavity of the spectral band.
... We now show how we select the channel lengths C e N 's of the target model enc. Following the rule of thumb [5,13,14,33,37] in PCA dimension reduction to keep the most important information, the target layer reluN 1 e of the channel length C e N should preserve 85% of the variance information from the source layer reluN 1. The channel length that preserves this amount of information can be determined with a data-driven approach. ...
Photorealistic style transfer entails transferring the style of a reference image to another image so the result seems like a plausible photo. Our work is inspired by the observation that existing models are slow due to their large sizes. We introduce PCA-based knowledge distillation to distill lightweight models and show it is motivated by theory. To our knowledge, this is the first knowledge distillation method for photorealistic style transfer. Our experiments demonstrate its versatility for use with different backbone architectures, VGG and MobileNet, across six image resolutions. Compared to existing models, our top-performing model runs at speeds 5-20x faster using at most 1\% of the parameters. Additionally, our distilled models achieve a better balance between stylization strength and content preservation than existing models. To support reproducing our method and models, we share the code at \textit{https://github.com/chiutaiyin/PCA-Knowledge-Distillation}.
... The results obtained by [59,60] in mapping elements for the characterization of the soil (C, clay and pH), captured using spectroscopy techniques and analyzed with the PLSR method, have been able to zone subareas with great heterogeneity among them, with very high R 2 values (0.92-0.93). The authors of [61] have also studied vineyards with differentiated subzones, with respect to the characterization of the vineyard soils, in this case, in relation with the topography parameters. ...
The characterization of vineyard soil is a key issue for crop management, which directly affects the quality and yield of grapes. However, traditional laboratory analysis of soil properties is tedious and both time and cost consuming, which is not suitable for precision viticulture. For this reason, a fast and convenient soil characterization technique is needed for soil quality assessment and precision soil management. Here, spectroscopy appears as a suitable alternative to assist laboratory analysis. This work focuses on estimating soil properties by spectroscopy. Our study was carried out using 96 soil samples collected from three vineyards in Rias Baixas Designation of Origen (Galicia, Spain). The soils that were characterized include nitrogen (N), organic matter (OM) and clay content (Clay). The presented work compared two regression techniques (partial least squares (PLSR) and random forest (RF)) and four spectral ranges: visible—VIS (350–700 nm), near infrared—NIR (701–1000 nm), short wave infrared—SWIR (1001–2500 nm) and VIS-NIR-SWIR (350–2500 nm) in order to identify the more suitable prediction models. Moreover, the effect of pre-treatments in reflectance data (smoothing Svitzky–Golay, SG, baseline normalization, BN, first derivative, FD, standard normal variate, SNV, logarithm of 1/reflectance or spectroscopy (SP) and detrending, SNV-D) was evaluated. Finally, continuous maps of the soil properties were created based on estimated values of regression models. Our results identified PLSR as the best regression technique, with less computation time than RF. The data improved after applying transformation in reflectance data, with the best results from spectroscopy pre-treatment (logarithm of 1/Reflectance). PLSR performances have obtained determination coefficients (R2) of 0.69, 0.73 and 0.52 for nitrogen, organic matter, and clay, respectively, with acceptable accuracy (RMSE: 0.03, 1.06 and 2.90 %) in a short time. Furthermore, the mapping of soil vineyards generates information of high interest for the precision viticulture management, as well as a comparison between the methodologies used.
... Most of the studies required a large dataset of soil samples to provide sufficient input for calibration and validation, which is unfavourable under practical considerations (field work, laboratory work and costs) [16,26]. Hence, the need of an efficient sampling strategy to avoid pseudo-replications and selecting truly representative locations for an accurate prediction of SOC have been emphasized [17]. ...
Remote sensing plays an increasingly key role in the determination of soil organic carbon (SOC) stored in agriculturally managed topsoils at the regional and field scales. Contemporary Unmanned Aerial Systems (UAS) carrying low-cost and lightweight multispectral sensors provide high spatial resolution imagery (<10 cm). These capabilities allow integrate of UAS-derived soil data and maps into digitalized workflows for sustainable agriculture. However, the common situation of scarce soil data at field scale might be an obstacle for accurate digital soil mapping. In our case study we tested a fixed-wing UAS equipped with visible and near infrared (VIS-NIR) sensors to estimate topsoil SOC distribution at two fields under the constraint of limited sampling points, which were selected by pedological knowledge. They represent all releva nt soil types along an erosion-deposition gradient; hence, the full feature space in terms of topsoils’ SOC status. We included the Topographic Position Index (TPI) as a co-variate for SOC prediction. Our study was performed in a soil landscape of hummocky ground moraines, which represent a significant of global arable land. Herein, small scale soil variability is mainly driven by tillage erosion which, in turn, is strongly dependent on topography. Relationships between SOC, TPI and spectral information were tested by Multiple Linear Regression (MLR) using: (i) single field data (local approach) and (ii) data from both fields (pooled approach). The highest prediction performance determined by a leave-one-out-cross-validation (LOOCV) was obtained for the models using the reflectance at 570 nm in conjunction with the TPI as explanatory variables for the local approach (coefficient of determination (R²) = 0.91; root mean square error (RMSE) = 0.11% and R² = 0.48; RMSE = 0.33, respectively). The local MLR models developed with both reflectance and TPI using values from all points showed high correlations and low prediction errors for SOC content (R² = 0.88, RMSE = 0.07%; R² = 0.79, RMSE = 0.06%, respectively). The comparison with an enlarged dataset consisting of all points from both fields (pooled approach) showed no improvement of the prediction accuracy but yielded decreased prediction errors. Lastly, the local MLR models were applied to the data of the respective other field to evaluate the cross-field prediction ability. The spatial SOC pattern generally remains unaffected on both fields; differences, however, occur concerning the predicted SOC level. Our results indicate a high potential of the combination of UAS-based remote sensing and environmental covariates, such as terrain attributes, for the prediction of topsoil SOC content at the field scale. The temporal flexibility of UAS offer the opportunity to optimize flight conditions including weather and soil surface status (plant cover or residuals, moisture and roughness) which, otherwise, might obscure the relationship between spectral data and SOC content. Pedologically targeted selection of soil samples for model development appears to be the key for an efficient and effective prediction even with a small dataset.
... Some researchers have noted that this technique can estimate primary soil properties (such as total C, N, and exchange capacity), as well as secondary soil properties (such as the respiration rate and potentially mineralizable N) at the same time [1][2][3][4][5][6]. Other authors have described the application of additional physical and soil chemical properties [7][8][9][10][11][12]. ...
Near-infrared reflectance spectroscopy (NIRS) was successfully used in this study to measure soil properties, mainly C and N, requiring spectral pre-treatments. Calculations in this evaluation were carried out using multivariate statistical procedures with preceding pre-treatment procedures of the spectral data. Such transformations could remove noise, highlight features, and extract essential wavelengths for quantitative predictions. This frequently significantly improved the predictions. Since selecting the appropriate transformation was not straightforward due to the large numbers of available methods, more comprehensive insight into choosing appropriate and optimized pre-treatments was required. Therefore, the objectives of this study were (i) to compare various pre-processing transformations of spectral data to determine their suitability for modeling soil C and N using NIR spectra (55 pre-treatment procedures were tested), and (ii) to determine which wavelengths were most important for the prediction of C and N. The investigations were carried out on an arable field in South Germany with a soil type of Calcaric Fluvic Relictigleyic Phaeozem (Epigeoabruptic and Pantoclayic), created in the flooding area of the Isar River. The best fit and highest model accuracy for the C (Ct, Corg, and Ccarb) and N models in the calibration and validation modes were achieved using derivations with Savitzky–Golay (SG). This enabled us to calculate the Ct, Corg, and N with an R2 higher than 0.98/0.86 and an ratio of performance to the interquartile range (RPIQ) higher than 10.9/4.1 (calibration/validation).
... However, it was not possible within this sequential fractionation to estimate successfully the contents of stable P o in the same concentrated HCl extract (Negassa and Leinweber, 2009) by MIRS and PLSr. Nevertheless, further progress in this research is expected due to advances in spectrometer hardand software, and more effective mathematical spectra analysis that is based on support vector machines, wavelets and/or neural networks (e.g., Daniel et al., 2003;Odlare et al., 2005). For example, Daniel et al. (2003) used neural networks to estimate P from the VIS-NIR spectrum. ...
... First, the PCA were used due to significance role in soil applications (Odlare et al. 2005;Viscarra Rossel et al. 2006;Viscarra Rossel et al. 2009;Savvides et al. 2010). PCA is a linear transformation technique related to Factor Analysis, and has been used in remote sensing to produce uncorrelated output bands, to segregate noise components, and to reduce the spectral dimensionality of data (Richards and Jia 2006). ...
The study anticipated to understand sand encroachment evolution through analysis of sand contribution across space and time using remote sensing in Laâyoune-Tarfaya basin, Morocco, over the period from 1987 to 2011. The assessment based on supervised classifications of Landsat imagery orthorectified data, using Maximum Likelihood (ML), Minimum Distance (MD), and Support Vector Machine (SVM) classifiers.
In order to ameliorate the information, principal components analysis (PCA) and co-occurrence measurement algorithm were used for choosing bands and data transformation. Images differencing was applied on image pairs derived from classification to analyze sand encroachment evolution. All classifiers present enhanced performances, and revealed that area covered by sand was increased by 7%, 4.66%, and 4.59% for ML, MD, and SVM respectively. Consequently, images differencing results confirmed that sand material increasing arise not only from coastal area contribution but mostly from erosion of complicated sand dunes exist in the middle part of the studied area.
Keywords: Maximum Likelihood, Support Vector Machine, Images differencing, coastal areas, sand dunes
... The NIR spectrum of a soil sample contains information about its physical, chemical and biological properties (Guerrero et al., 2016). In the relevant region of the electromagnetic spectrum, each constituent of a complex organic mixture (C, N, H and O) has unique absorption properties because of stretching and bending vibrations in molecular bands (Odlare et al., 2005). These encode information on soil properties and are comparable to a fingerprint of the sample. ...
Research has shown that the application of near‐infrared (NIR) spectroscopy can be used to predict soil attributes, in particular for regional to continental scales. However, there are challenges when NIR is used at the regional scale because of the considerable spatial variation. This study has predicted SOC at the country scale (German agricultural soil inventory) with different stratification strategies for NIR data: (i) calibration with memory‐based learning (MBL) algorithms that use spectral similarity and (ii) simple stratification based on soil properties (depth, pH and soil texture) and land use. To optimize calibration models, this study aimed to predict soil organic carbon (SOC) determined by these three strategies for 1410 soil profiles selected from the German agricultural soil inventory. The profiles covered a wide range of soil types and characteristics. The calibration procedures were based on complete soil profile data of two‐thirds of the dataset and one‐third of the dataset was used for independent validation (prediction); the profiles were selected randomly. Available soil properties for stratifying the datasets were: soil depth (topsoil 0–30 cm and subsoil 31–100 cm), pH and texture class (silty, clayey, sandy and loamy). The profiles were also stratified by land use (cropland and grassland) and with the MBL method. The calibrations were carried out by partial least‐squares regression (PLSR), and each stratification model was compared with the global model. The root mean square error of cross‐validation (RMSECV) for the global model was 4.2 g SOC kg⁻¹. Stratification according to soil depth reduced the error by 10% (RMSECV 3.8 g kg⁻¹). The best stratification by soil texture was when sandy soil samples were separated from the other samples, which reduced the RMSECV by 14%. Calibration with MBL provided the most accurate predictions of SOC, with an error reduction of 25% (RMSECV 3.2 g kg⁻¹). Thus, calibrations with NIR of country‐scale datasets can be improved easily by stratification or application of the MBL algorithm.
Highlights
• Large country‐scale soil dataset of near‐infrared with > 1400 soil profiles was used.
• Stratification of NIR data by soil properties increased the accuracy of SOC calibration at country scale.
• The best stratification design involved calibrating sandy soil separately from other texture classes.
• A decrease of 25% in calibration error and 22% in prediction error with the MBL model compared with global model.
... Many researchers working on this approach have demonstrated that the optical method is capable of providing low-cost, rapid and reliable products compared to the intensive laboratory work (Arslan et al., 2014;Christy, 2008;Cohen et al., 2005;Dunn et al., 2002;Janik et al., 2009;Reeves and Smith, 2009;Viscarra Rossel et al., 2009;Viscarra Rossel and Webster, 2012;Volkan Bilgili et al., 2010;Zornoza et al., 2008). A number of studies have shown the potential of VNIR-SWIR data analysis using principal component analysis (Odlare et al., 2005;Rizzo and Demattê, 2014;Verheyen et al., 2001) and fuzzy clustering algorithms (Bezdek et al., 1984;De Gruijter et al., 1997;McBratney and de Gruijter, 1992;Odeh et al., 1990;Odgers et al., 2011aOdgers et al., , 2011bVerheyen et al., 2001). Other methods, such as increasing sampling density (Wetterlind et al., 2010) and GIS techniques for modeling soilgenerating factors (Karnieli et al., 1998), have proven to be reliable for soil classification based on the relationship between soil properties and soil types. ...
Visible, near-infrared and shortwave-infrared (VNIR–SWIR) spectroscopy has proven to be an efficient, rapid and low-cost method for soil spectral analysis that can improve on the results obtained from today's traditional methods of conducting soil surveys. Nonetheless, this tool is used mostly in the laboratory and at surface level. The main objective of this paper is to develop a new optical method for characterizing soil profiles, towards improving the efficiency and accuracy of the traditional soil survey. We used airborne hyperspectral data from the AisaFENIX sensor for surface classification and ASD spectral measurements of soil samples for subsurface analysis. A total of 643 soil samples were extracted from 48 cores, each core representing a soil profile. All samples were air-dried, crushed and sieved, and then analyzed by ASD spectrometer under laboratory conditions. Clay content was also measured to provide additional information. The 3D spectral data were analyzed using SAM algorithm, spectral gradient (m), k-means clustering and gley horizon parameter (G) to classify soils and distinguish between soil horizons in each core. The results suggest that these parameters can provide satisfactory results both from laboratory measurements and hyperspectral remote sensing data (R 2 = 0.81 for clay content and R 2 = 0.78 for gleying conditions) in order to distinguish between the soil horizons using 3D spectral information. Moreover, the method is satisfactory for obtaining soil types from 3D spectral sensing as well as evaluating the catena development and other spatial soil distributions.
... Many researchers working on this approach have demonstrated that the optical method is capable of providing low-cost, rapid and reliable products compared to the intensive laboratory work (Arslan et al., 2014;Christy, 2008;Cohen et al., 2005;Dunn et al., 2002;Janik et al., 2009;Reeves and Smith, 2009;Viscarra Rossel et al., 2009;Viscarra Rossel and Webster, 2012;Volkan Bilgili et al., 2010;Zornoza et al., 2008). A number of studies have shown the potential of VNIR-SWIR data analysis using principal component analysis (Odlare et al., 2005;Rizzo and Demattê, 2014;Verheyen et al., 2001) and fuzzy clustering algorithms (Bezdek et al., 1984;De Gruijter et al., 1997;McBratney and de Gruijter, 1992;Odeh et al., 1990;Odgers et al., 2011aOdgers et al., , 2011bVerheyen et al., 2001). Other methods, such as increasing sampling density (Wetterlind et al., 2010) and GIS techniques for modeling soilgenerating factors (Karnieli et al., 1998), have proven to be reliable for soil classification based on the relationship between soil properties and soil types. ...
Visible, near-infrared and shortwave-infrared (VNIR–SWIR) spectroscopy has proven to be an efficient, rapid and low-cost method for soil spectral analysis that can improve on the results obtained from today's traditional methods of conducting soil surveys. Nonetheless, this tool is used mostly in the laboratory and at surface level. The main objective of this paper is to develop a new optical method for characterizing soil profiles, towards improving the efficiency and accuracy of the traditional soil survey. We used airborne hyperspectral data from the AisaFENIX sensor for surface classification and ASD spectral measurements of soil samples for subsurface analysis. A total of 643 soil samples were extracted from 48 cores, each core representing a soil profile. All samples were air-dried, crushed and sieved, and then analyzed by ASD spectrometer under laboratory conditions. Clay content was also measured to provide additional information. The 3D spectral data were analyzed using SAM algorithm, spectral gradient (m), k-means clustering and gley horizon parameter (G) to classify soils and distinguish between soil horizons in each core. The results suggest that these parameters can provide satisfactory results both from laboratory measurements and hyperspectral remote sensing data (R 2 = 0.81 for clay content and R 2 = 0.78 for gleying conditions) in order to distinguish between the soil horizons using 3D spectral information. Moreover, the method is satisfactory for obtaining soil types from 3D spectral sensing as well as evaluating the catena development and other spatial soil distributions.
... Investigations to analyze agroecosystem properties using visible-, near-, and mid-infrared spectroscopy jointly with geostatistics (e.g., for soil properties) (Odlare et al., 2005;Cobo et al., 2010;Bilgili et al., 2011;Conforti et al., 2013;Shen et al., 2013;Steffens and Buddenbaum, 2013) can be done: (i) in situ, (ii) with samples in the laboratory, or (iii) from images collected by satellites or unmanned aerial vehicles. Linking spectroscopy with geostatistics can be done by (i) processing spectral information via chemometrics and using the predicted results in spatial analyses or (ii) through integration of chemometric coefficients [e.g., partial least squares (PLS) scores] to improve the modeling of the geostatistical step. ...
Core Ideas
Near‐infrared spectroscopy (NIRS) was appropriate for predicting soil texture and C.
Both linear and nonlinear multivariate models could be used for NIRS calibration.
Soil texture was predicted with greater precision than organic C fractions.
Near‐infrared spectroscopy and kriging were a useful combination for assessing spatial variation.
Near‐infrared spectroscopy (NIRS) and geostatistics are relatively unexplored tools that could reduce the time, labor, and costs of soil analysis. Our objective was to efficiently determine lateral and vertical distributions of soil texture and soil organic C (SOC) fractions in an agroforestry system (a 7‐ha field) on a Coastal Plain site in North Carolina. To predict selected properties from a large number of soil samples collected from this field, NIRS was calibrated against laboratory‐determined properties. Support vector machines was a multivariate model that performed better than partial least squares to obtain greater precision with NIRS for all soil properties. To predict soil properties with precision across the field, geostatistical modeling with maximum likelihood and ordinary kriging was used. When we combined the two modeling processes, the root mean square error (RMSE) and the RMSE relative to the dataset mean (%RMSE) were 67 g kg ‐1 for sand (9.3% RMSE), 34 g kg ‐1 for clay (22.7% RMSE), 1.63 g kg ‐1 for total organic C (26.7% RMSE), 0.67 g kg ‐1 for particulate organic C (36.1% RMSE), and 24 mg CO 2 –C kg ‐1 3 d ‐1 for the flush of CO 2 (29% RMSE). We conclude that the combination of NIRS and kriging produced acceptable errors and therefore could be used to predict the spatial distribution of soil texture and SOC fractions in this agroforestry system to allow efficient assessment of management changes with time and better predict small‐scale input requirements.
... NIRS is a physically non-destructive, rapid, reproducible and low-cost method that characterizes materials according to their reflectance in the wavelength range between 800 and 2500 nm (Stevens et al., 2006;Brunet et al., 2007). In this region of the electromagnetic spectrum, each constituent of a complex organic mixture (C, N, H, O, P and S atoms) has unique absorption properties due to stretching and bending vibrations in molecular bonds (Odlare et al., 2005). The technique has been used in the past two decades to assess total carbon and nitrogen, nitrate-nitrogen (N-NO − 3 ), ammonia-nitrogen (N-NH + 4 ), respirometry, pH, cation exchange capacity, Ca, Mg, K, exchangeable Al, total P, electrical conductivity, particle size distribution, moisture, clay content and CaCO 3 (e.g. ...
Near infrared (NIR) and mid-infrared (mid-IR) reflectance spectroscopy are time- and cost-effective tools for characterising soil organic carbon (SOC). Here they were used for quantifying (i) carbon (C) dioxide (CO2) emission from soil samples crushed to 2 mm and 0.2 mm, at 18°C and 28°C; (ii) physical C protection, calculated as the difference between CO2 emissions from 0.2 mm and 2 mm crushed soil at a given temperature; and (iii) the temperature vulnerability of this protection, calculated as the difference between C protection at 18°C and 28°C. This was done for 97 topsoil samples from Tunisia, mostly calcareous, which were incubated for 21 days. Soil CO2 emission increased with temperature and fine crushing. However, C protection in 0.2–2 mm aggregates had little effect on the temperature vulnerability of CO2 emission, possibly due to preferential SOC protection in smaller aggregates. In general, NIR spectroscopy, and to a lesser extent mid-IR spectroscopy, yielded accurate predictions of soil CO2 emission (0.60 ≤ R2 ≤ 0.91), and acceptable predictions of C protection at the beginning of incubation (0.52 ≤ R2 ≤ 0.81) but not over the whole 21 day period (R2 ≤ 0.59). For CO2 emission, prediction error was the same order of magnitude as, and sometimes similar to, the uncertainty of conventional determination, indicating that a noticeable proportion of the former could be attributed to the latter. The temperature vulnerability of C protection could not be modelled correctly (R2 ≤ 0.11), due to error propagation. The prediction of SOC was better with NIR spectroscopy and that of soil inorganic C (SIC) was very accurate (R2 ≥ 0.94), especially with mid-IR spectroscopy. Soil CO2 emission, C protection and its vulnerability were best predicted with NIR spectra, those of 0.2 mm samples especially. With 2 mm samples, mid-IR spectroscopy yielded the worst predictions in general. NIR spectroscopy prediction models suggested that CO2 emission and C protection depended (i) on aliphatic compounds (i.e. labile substrates), dominantly at 18°C; (ii) on amides or proteins (i.e. microbial biomass), markedly at 28°C; and (iii) negatively, on organohalogens and aromatic amines (i.e. pesticides). Models using mid-IR spectra showed a negative influence of carbonates on CO2 emission, suggesting they did not contribute to soil CO2 emission and might form during incubation. They also suggested that CO2 emission and C protection related to carboxylic acids, saturated aliphatic ones especially.
... Predicted SOC showed nugget-sill ratios of 30.3% and 28.6%, respectively, and ranges of 1618 m and 1477 m, respectively. Nugget-sill ratios reported in the literature vary greatly depending on the local soil conditions and the sampling and analysis methods [32,61]. ...
With the upcoming availability of the next generation of high quality orbiting hyperspectral sensors, a major step toward improved regional soil mapping and monitoring and delivery of quantitative soil maps is expected. This study focuses on the determination of the prediction accuracy of spectral models for the mapping of common soil properties based on upcoming EnMAP (Environmental Mapping and Analysis Program) satellite data using semi-operational soil models. Iron oxide (Fed), clay, and soil organic carbon (SOC) content are predicted in test areas in Spain and Luxembourg based on a semi-automatic Partial-Least-Square (PLS) regression approach using airborne hyperspectral, simulated EnMAP, and soil chemical datasets. A variance contribution analysis, accounting for errors in the dependent variables, is used alongside classical error measurements. Results show that EnMAP allows predicting iron oxide, clay, and SOC with an R2 between 0.53 and 0.67 compared to Hyperspectral Mapper (HyMap)/Airborne Hyperspectral System (AHS) imagery with an R2 between 0.64 and 0.74. Although a slight decrease in soil prediction accuracy is observed at the spaceborne scale compared to the airborne scale, the decrease in accuracy is still reasonable. Furthermore, spatial distribution is coherent between the HyMap/AHS mapping and simulated EnMAP mapping as shown with a spatial structure analysis with a systematically lower semivariance at the EnMAP scale.
... The advantage of the NIR-PCA strategy is that the first PCs (PC 1) will capture the spectral bands that express the largest variation regardless of what the NIR bands correlate to and, hence, PC 1 will always explain the variation of the soil properties that in each specific case has the largest influence on the PCA model. NIR-PCA strategy seems to be an efficient and reliable strategy to use when determining the soil spatial variation in a field (Odlare et al., 2005). By using optical sensor and photo-multiplier tube, real-time in situ sensing of photosynthetic activity of plant has been done (Kebabian et al., 1998). ...
Precision agriculture (PA) concept was initiated for site specific crop management as a combination of positioning system technology, variable rate technology, remote sensing, yield mapping etc. to optimize the profitability, sustainability with a reduced environmental impact. From centuries Indian farms are experiencing some sort of soft precision agriculture technology. But the challenges of free and globalized market as well as ever-increasing population with huge food grain demand create the scope of adoption of hard precision agriculture technology in Indian farms. So learning the new agricultural technology invented in developed countries and its proper modification and application according to the domestic condition is necessary. Therefore, nearly hundred research papers generated in last three decades have been critically reviewed to find the status of main six components of PA, i.e.
... Existing studies have shown that VNIRS is available to provide rapid information about soil physical and chemical properties, e.g. moisture, carbon, nitrogen, phosphorus and calcium, and cation exchange capacity in an economical manner (Ben-Dor and Banin, 1995;Chang et al., 2001;Reeves and McCarty, 2001;Odlare et al., 2005;Cozzolino and Moron, 2006;Maleki et al., 2006;Wetterlind et al., 2008;Mouazen et al., 2009). However, many of the investigations involved a limited number of samples, or the samples came from a limited number of sites of similar range of soils. ...
... It is a simple, rapid, non-destructive, and cheap technique to handle a big set of samples as will be the case for the BZE samples where good quality, inexpensive soil data is required to monitor environmental changes. It has already been extensively used to determine SOC in the laboratory for the study of, for example, the spatial variability of SOC at the field level (Shepherd et al. 2002, Martin et al., 2002Odlare et al., 2005). Moreover a single spectrum may allow for characterisation of a number of soil properties. ...
... These methods, even when they are reliable, require 16 h to generate results and require high environmental impacts. Near-infrared spectroscopy (NIRS) has been utilized for predicting total concentrations and research note Analysis of soil organic matter in tropical soils with near-infrared spectroscopy (NIRS) and chemometrics Aarón Jarquín-Sánchez 1 , Sergio Salgado-García 2 , David J. Palma-López 2 , and Wilder Camacho-Chiu 3 the spatial variation of nitrogen and carbon (and other elements) in soils (Ludwing et al., 2002;Odlare et al., 2005;Shenk, 2004;Cozzolino and Moron, 2006). ...
The objective of this study was to predict the concentration of soil organic matter (SOM) in tropical soils using near-infrared spectroscopy (NIRS) for samples measured within a polyethylene bag and without a bag. One hundred and fifty six soil samples from the humid tropics of Tabasco, Mexico with contrasting chemical characteristics were selected. The samples were dried, ground, and sieved through 2 mm and 5 mm screens, and their SOM contents were determined using the Walkley-Black method. The soil samples were packed in polyethylene bags, and SOM was measured directly with and without a bag using a quartz probe (FOSS 5000 model of NIRsystems, DK-3400 Hillerod,Denmark) for a range from 1100 to 2000 nm to obtain a prediction model of SOM. The model for determining SOM for bagged samples had good fit and explained 88% of the variation (0 to 10, 2% of SOM of samples). The model for determining SOM for bag-less samples was not efficient for predicting independent samples and therefore was discarded. The analysis by NIRS was reliable, more rapid, and easier for the determination of SOM in soil samples measured through a plastic bag.
... Reflectance spectroscopy is used in chemometrics to construct spectral groups' classification and regression models to predict many soil attributes. While regression methods are used to model the spectral signature of a target based on specific physical, biological, or chemical soil properties, classification is used to group the spectral signatures of soil into categories [24][25][26][27]. The statistical models have included parametrical methods such as partial least squares-regression (PLS-R), which is perhaps the most commonly used regression method technique (e.g., [12,14,28]). ...
Soil quality (SQ) assessment has numerous applications for managing sustainable soil function. Airborne imaging spectroscopy (IS) is an advanced tool for studying natural and artificial materials, in general, and soil properties, in particular. The primary goal of this research was to prove and demonstrate the ability of IS to evaluate soil properties and quality across anthropogenically induced land-use changes. This aim was fulfilled by developing and implementing a spectral soil quality index (SSQI) using IS obtained by a laboratory and field spectrometer (point scale) as well as by airborne hyperspectral imaging (local scale), in two experimental sites located in Israel and Germany. In this regard, 13 soil physical, biological, and chemical properties and their derived soil quality index (SQI) were measured. Several mathematical/statistical procedures, consisting of a series of operations, including a principal component analysis (PCA), a partial least squares-regression (PLS-R), and a partial least squares-discriminate analysis (PLS-DA), were used. Correlations between the laboratory spectral values and the calculated SQI coefficient of determination (R2) and ratio of performance to deviation (RPD) were R2 = 0.84; RPD = 2.43 and R2 = 0.78; RPD = 2.10 in the Israeli and the German study sites, respectively. The PLS-DA model that was used to develop the SSQI showed high classification accuracy in both sites (from laboratory, field, and imaging spectroscopy). The correlations between the SSQI and the SQI were R2 = 0.71 and R2 = 0.7, in the Israeli and the German study sites, respectively. It is concluded that soil quality can be effectively monitored using the spectral-spatial information provided by the IS technology. IS-based classification of soils can provide the basis for a spatially explicit and quantitative approach for monitoring SQ and function at a local scale.
... Using infrared spectroscopy combined with Geographic Information System (GIS) and statistical methods, the N, P, K, and soil organic matter (SOM) spatial variability within the field can be obtained (Odlare et al. 2005;Christy 2008;Wetterlind et al. 2008a), and their distribution maps can be drawn (He et al. 2005) in which soil nutrient status can be directly indicated. The reference maps for the predicted and measured values of N and OM were almost the same, unlike with P and K, due to the unsuccessful prediction of these constituents. ...
... Los mayores valores de correlación (valor absoluto) encontrados en la primera parte de las longitudes de onda en la región NIR para CT, Fe y arcilla, confirman una vez más la incidencia de estas propiedades en el comportamiento de la respuesta espectral. Altas correlaciones entre estas propiedades y diferentes longitudes de onda son también reportadas por diferentes investigadores (Leone & Sommer, 2000;McCarty et al., 2002;Mouazen et al., 2007;Odlare et al., 2005;Santra et al., 2009;Vohland et al.;Volkan Bilgili et al., 2010). Por otra parte, propiedades como P, Ca, Na y Zn, presentan valores bajos de correlación en todas las longitudes de onda de la región NIR, coincidiendo con la no apreciación de correlaciones lineales significativas analizadas anteriormente. ...
The characterization of soil properties using laboratory tests is a fundamental part in the diagnosis of potential land use and fertility. The traditional physical and chemical analyzes are expensive and time consuming, hampering the adoption of crop management technologies, such as precision agriculture. The aim of this study was to determine the physical and chemical properties of a Typic Hapludox using NIR diffuse reflectance spectroscopy and conventional methods. Using a rigid grid system, 1240 samples were collected in the A and B horizons soils, to determine the contents of total carbon (CT), total nitrogen (NT), pH, exchangeable acidity (Ac.I), exchangeable aluminum (Al.I ) cation exchange capacity (CEC), P, Ca, Mg, K, Na, Cu, Fe, Mn, Zn, sand, silt and clay. Also obtained NIR spectral signatures of each sample and models were developed by partial least squares regression, with different set size samples. The use of diffuse reflectance spectroscopy and statistical methods allowed the quantification of ten analyzed properties (CT, NT, Cu, Fe, Clay, Mn, Ac.I, Al.I, K, CEC). For CT, Cu and Fe, it was possible to obtain adequate prediction models with low calibration set size (less than 150) of the total, RPD greater than 2.0, R2 greater than 0.80 and lower RMSE. The results obtained from NIR models could be directly integrated into geostatistical evaluations, obtaining similar digital and spectrum digital maps, to those properties with representative models. The use of pedometrics methods has shown promising results for these soils and constitutes a basis for the development of this soil science research area in Colombia
... However, it was not possible within this sequential fractionation to estimate successfully the contents of stable P o in the same concentrated HCl extract (Negassa and Leinweber, 2009) by MIRS and PLSr. Nevertheless, further progress in this research is expected due to advances in spectrometer hardand software, and more effective mathematical spectra analysis that is based on support vector machines, wavelets and/or neural networks (e.g., Daniel et al., 2003;Odlare et al., 2005). For example, Daniel et al. (2003) used neural networks to estimate P from the VIS-NIR spectrum. ...
... The soil was analysed for chemical, physical and microbiological properties before the start of the experiment (Odlare et al. 2005(Odlare et al. , 2008. The results generally had small coefficients of variance, indicating a low spatial variation at the experimental site. ...
An 8-year-long field experiment (1998–2006) was established in Sweden with the aim of evaluating the effects of applying organic wastes in combination with mineral nitrogen (N) to agricultural soil. Sewage sludge (SS), biogas residues (BR) and municipal compost (CO) were applied annually at rates corresponding to 50 kg N/ha and supplementary mineral N fertilizer also applied at rates corresponding to 50 kg N/ha. The effects were evaluated by analysing crop yield and soil chemical and microbiological properties. The results showed that none of the fertilizers produced significantly higher yield of barley over the 8-year period compared to any other. Biogas residue proved to be particularly beneficial for the substrate-induced respiration (SIR) in soil and increased the proportion of active to dormant micro-organisms. Treatment with SS increased plant-available phosphorus (P-AL) and N mineralization (N-min), whereas CO increased the basal respiration (B-resp). Changes in the microbial community structure were assayed by terminal restriction fragment length polymorphism (T-RFLP); the T-RFLP signatures of the soil bacterial community were largely unaffected by the addition of organic waste. Of the chemical properties assayed, the largest increases were seen in P-AL, where SS produced the highest value. Treatments with the organic wastes showed no negative effects other than a slight decrease in B-resp induced by SS and BR. In conclusion, the microbiological activity in the soil responded more rapidly than the changes in the community structure and the chemical properties to changes in the soil environment.
... However, it was not possible within this sequential fractionation to estimate successfully the contents of stable P o in the same concentrated HCl extract (Negassa and Leinweber, 2009) by MIRS and PLSr. Nevertheless, further progress in this research is expected due to advances in spectrometer hardand software, and more effective mathematical spectra analysis that is based on support vector machines, wavelets and/or neural networks (e.g., Daniel et al., 2003;Odlare et al., 2005). For example, Daniel et al. (2003) used neural networks to estimate P from the VIS-NIR spectrum. ...
Available open access: http://onlinelibrary.wiley.com/doi/10.1002/jpln.201400327/abstract
Phosphorus (P) is an indispensable element for all life on Earth and, during the past decade, concerns about the future of its global supply have stimulated much research on soil P and method development. This review provides an overview of advanced state-of-the-art methods currently used in soil P research. These involve bulk and spatially resolved spectroscopic and spectrometric P speciation methods (1 and 2D NMR, IR, Raman, Q-TOF MS/MS, high resolution-MS, NanoSIMS, XRF, XPS, (µ)XAS) as well as methods for assessing soil P reactions (sorption isotherms, quantum-chemical modeling, microbial biomass P, enzymes activity, DGT, 33P isotopic exchange, 18O isotope ratios). Required experimental set-ups and the potentials and limitations of individual methods present a guide for the selection of most suitable methods or combinations.
... However, such soil analyses are laborious, costly, time consuming, and inadequate when high spatial and temporal resolution of TPH content are warranted. 7 Consequently, there is a persistent need for the development of innovative, low-cost, and reproducible analytical package for mapping spatial variability of petroleum contaminated soils. ...
Visible near-infrared (VisNIR) diffuse reflectance spectroscopy (DRS) is a rapid, non-destructive method for sensing the presence and amount of total petroleum hydrocarbon (TPH) contamination in soil. This study demonstrates the feasibility of VisNIR DRS to be used in the field to proximally sense and then map the areal extent of TPH contamination in soil. More specifically, we evaluated whether a combination of two methods, penalized spline regression and geostatistics could provide an efficient approach to assess spatial variability of soil TPH using VisNIR DRS data from soils collected from an 80 ha crude oil spill in central Louisiana, USA. Initially, a penalized spline model was calibrated to predict TPH contamination in soil by combining lab TPH values of 46 contaminated and uncontaminated soil samples and the first-derivative of VisNIR reflectance spectra of these samples. The r 2 , RMSE, and bias of the calibrated penalized spline model were 0.81, 0.289 log 10 mg kg À1 , and 0.010 log 10 mg kg À1 , respectively. Subsequently, the penalized spline model was used to predict soil TPH content for 128 soil samples collected over the 80 ha study site. When assessed with a randomly chosen validation subset (n ¼ 10) from the 128 samples, the penalized spline model performed satisfactorily (r 2 ¼ 0.70; residual prediction deviation ¼ 2.0). The same validation subset was used to assess point kriging interpolation after the remaining 118 predictions were used to produce an experimental semivariogram and map. The experimental semivariogram was fitted with an exponential model which revealed strong spatial dependence among soil TPH [r 2 ¼ 0.76, nugget ¼ 0.001 (log 10 mg kg À1) 2 , and sill 1.044 (log 10 mg kg À1) 2 ]. Kriging interpolation adequately interpolated TPH with r 2 and RMSE values of 0.88 and 0.312 log 10 mg kg À1 , respectively. Furthermore, in the kriged map, TPH distribution matched with the expected TPH variability of the study site. Since the combined use of VisNIR prediction and geostatistics was promising to identify the spatial patterns of TPH contamination in soils, future research is warranted to evaluate the approach for mapping spatial variability of petroleum contaminated soils.
... Within the geostatistical framework it is possible to integrate Vis-NIR and PXRF measurements. Odlare et al. (2005) and Cobo et al. (2010) used variograms produced with near-and mid-infrared reflectance spectroscopy data to study spatial variation in soil properties. More advanced applications to assess soil contamination using proximal sensing data can be provided using co-kriging or co-simulation. ...
There are tens of millions of contaminated soil sites in the world, and with an increasing population and associated risk there is a growing pressure to remediate them. A barrier to remediation is the lack of cost-effective approaches to assessment. Soil contaminants include a wide range of natural and synthetic metallic and organic compounds and minerals thus making analytical costs potentially very large. Further, soil contaminants show a large degree of spatial variation which increases the burden on sampling costs. This paper reviews potentially cost-effective methods for measurement, sampling design, and assessment. Current tiered investigation approaches and sampling strategies can be improved by using new technologies such as proximal sensing. Design of sampling can be aided by on-the-go proximal soil sensing; and expedited by subsequent adaptive spatially optimal sampling and prediction procedures enabled by field spectroscopic methods and advanced geostatistics. Field deployment of portable Visible & Near Infrared [wavelength 400–2500 nm] (Vis-NIR) and X-ray fluorescence (PXRF) spectroscopies will require special calibration approaches but show huge potential for synergistic use. The use of mid-infrared spectroscopy [wavelength 2500–25,000 nm, wavenumber 4000–400 cm− 1] (MIR) for field implementation requires further adaptive research. We propose an integrated field-deployable methodology as a basis for further developments.
Realizing the goal of food security calls for more and sustained crop production per unit of land. Food security could be achieved by improving soil productivity. In order to improve soil productivity timely and reliable information on soil fertility is a prerequisite.
This paper presents a literature review on the development of near infrared reflectance spectroscopy for soil analysis and the contribution of this technique to the evaluation of soil fertility analysis. This technique is used to estimate the chemical composition of soil samples on the basis of their absorption properties. It is therefore an indirect method of measurement, which requires a calibration phase for the prediction of these properties. NIR spectroscopy offers many advantages compared to reference analysis: it is known to be a physical, non-destructive, rapid, reproducible and low cost method. Often employed in other analytical domains, such as agro-food, NIR spectroscopy has, however, seldom been used in soil characterization, due to the complexity of the soil matrix. Thanks to the development of chemometric methods, numerous studies have recently been conducted to evaluate the feasibility of the application of the technique in soil analysis. Most authors conclude that NIR spectroscopy is promising; however, to date, use of the technique has not spread to routine laboratories. The paper is organized as follows. Firstly, we provide an overview of the NIR spectroscopy technique and related chemometric methods. Secondly, we describe the soil characteristics that can be predicted using this technique. Finally, we detail examples of results that have been obtained through the use of the technique, mainly in the determination of clay and organic carbon content, and of cation exchange capacity.
Environmental studies often require analyses of numerous chemical, physical and biological properties in large numbers of soil, litter and plant samples. Such analyses may be expensive and time consuming and therefore rapid and cost-effective methods may be required. Near infrared spectroscopy (N1RS) is a non-destructive analytical method known for rapidity, simplicity and low costs, which could be used along with classical analytical methods in order to improve efficiency of large-scale environmental research. In this review, principals of NIRS are described, examples of NIRS applications are presented and the possibilities and limitations of the method are discussed.
To reduce and recycle biowaste volumes, many cities have chosen to build municipal composting plants. Presently, compost quality determination requires many biological and chemical characterizations. Aims of this study were to characterize composting processes and to develop a simple and valuable method to assess biological and chemical evolution of composts. Compost windrows obtained from municipal solid wastes mixed to green wastes, were matured during six months. Composts were characterized by chemical (elemental analysis, organic matter, lignin, humic substances and 13C NMR) and biological methods (respiration, micro-organism enumeration, metabolic profile modifications of microbial communities, enzymatic activities) and revealed mineralization and humification of organic matter. Efficiency of Near InfraRed Spectroscopy (NIRS) was then demonstrated with help of a data bank establishing from 432 composts. Finally, since no parameter can be used alone, a global index of composting evolution (GICE) was synthesized by PCA from the whole of information and was perfectly calibrated by NIRS with PLS method. GICE should be a simple and efficiency tool of maturity determination. This NIRS-PCA strategy could be applied to the optimization of a new composting process in reactor which is in development.
Effective protection of environmental resources requires an understanding of spatial and temporalvariation of soil properties, and its causes and its consequences. However, the acquisition of data on thisvariability requires datasets whose achievement is limited by the cost of physico-chemical analyses, and also insome cases by the amount of material available. For studies related to carbon sequestration in soils and theirvulnerability to land use and climate changes, additional limitations appear in carbonate soils as organic formsand inorganic carbon must be distinguished. Assuming that the near-infrared spectroscopy (NIRS) could helpovercome these limitations, this paper presents some results obtained with NIRS through two examples ofstudies in carbonate environment, one in the watershed the Rheraya (High Atlas of Morocco) and the other inthe plain of Languedoc (France). These examples show that depending on the particular case studied andphysicochemical methods implemented a procedure for prediction by NIRS may be efficient to predict the Corgand inefficient for predicting carbonates or moderately efficient in predicting the Corg and very efficient to predictthe carbonates.
http://eusoils.jrc.it/ESDB_Archive/eusoils_docs/other/EUR23290.pdf
ZET:Hassas tarım tekniklerinin uygulanması, küresel olarak toprakta karbon zenginleşmesinin gözlemlenmesi ve toprak kalitesinin sürdürülebilirliğini sağlayacak toprak özelliklerinin daha hızlı belirlenebileceği, ucuz ve güvenilir yöntemlere olan gereksinim sürekli artmaktadır. Toprakların fiziksel, kimyasal, biyolojik ve mineralojik özelliklerinin mevcut laboratuar yöntemler ile belirlenmesi pahalı ve oldukça zaman ve işçilik gerektirdiği gibi, analiz için kullanılan güçlü kimyasalların atıkları çevreye zarar verebilmektedir. Geleneksel olarak kullanılan laboratuar yöntemlerine alternatif olarak son zamanlarda yaygın bir şekilde kullanılmaya başlanan dağılmış yansıma spektroskopi (morötesi, görülebilir, yakın kızıl ötesi ve orta kızıl ötesi) tekniği, pH, organik karbon, su içeriği, parçacık büyüklük dağılımı, katyon değişim kapasitesi, değişebilir katyonlar, kil mineralojisi ve daha bir çok toprak özelliğinin hızlı bir şekilde belirlenmesine olanak vermektedir. Yansıma özelliklerinden gidilerek toprak özelliklerinin belirlenmesinde gelişmiş istatistiksel yöntemlerden faydalanılmaktadır. Çoklu regresyon analizi, temel bileşenler analizi, kısmi en az-karelerin regresyonu ve sinir ağları kalibrasyonu yaygın olarak kullanılan yöntemlerdir. Toprak özelliklerinin bozulmadan, yerinde incelenebilmesine olanak veren taşınabilir spektroskopi cihazları, arazideki değişkenliğin daha güvenilir şekilde incelenmesine olanak tanımaktadır. Tüm bu avantajlarının yanında, spektroskopik yöntemlerin doğruluğu kalibrasyona ve kullanılan referans metodun hassasiyeti ve doğruluğuna oldukça bağlıdır. Bu nedenle aletin kalibrasyonunda doğruluğu kabul edilmiş olan referans metotların kullanılması kaçınılmazdır. Anahtar Kelimeler: Yakın kızıl ötesi, NIRS, Toprak özellikleri,Toprak analizleri ABSTRACT:Application of precision agriculture techniques, monitoring of carbon sequestration in soils throughout the world and sustaining the soil quality require reliable, fast and cheap soil analysis techniques. The determination of soil physical, chemical, biological and mineralogical characteristics with conventional laboratory analysis can be great time and labor consuming and expensive. The waste of strong chemicals used in soil analysis is hazardous to environment. Diffuse reflectance spectroscopy (ultraviolet, visible, near infrared, mid infrared) as an alternative to conventional laboratory methods has been recently used to determine soil characteristics (soil pH, organic carbon, water content, particle size distribution, cation exchange capacity, exchangeable cations, clay mineralogy and many others) rapidly and inexpensively. Spectroscopic methods require the development of calibrations that relate the spectral information to the property of interest using several statistical methods. Multiple regression analysis, principal component analysis, partial least square regression and neural network are the commonly used multivariate statistical procedures. Portable spectroscopy equipments allow in situ characterization of soil characteristics; thereby variability of soil properties can be also determined in the field. Furthermore, the accuracy of spectroscopic techniques depends on the calibration and the precision and accuracy of the reference method. Therefore, reliable analytical methods need to be used in calibration of spectroscopic technique used in the analysis.
A fertility diagnosis, as close as possible to field truth, needs the knowledge of the cation exchange capacity (CEC) or, at least, the clay content of the composite samples. But, in the Walloon region (Belgium), these parameters are rarely analysed in routine and therefore, are appreciated on the field or estimated by regional average.
To overcome this gap, we investigated the use of near infrared spectroscopy (NIR) for the determination of three parameters.: CEC, total organic carbon and clay content.
900 samples representative of the Walloon region diversity were analysed. The CEC, clay content and total organic carbon, on air-dried samples, sieved in 2mm, were measured by reference methods (Metson, chain Hydrometer, Springler-Klee method) and by NIR.
The first stages of calibration, combined with morphopedologic stratifications, show very encouraging results. The accuracy of the predictions is sufficient for use within the framework of fertility evaluation.
Indeed, NIR appears as a really easy method which would allow analysis without extra cost to the farmer or the laboratory providing the diagnosis and advice.
Advanced classifiers, e.g., partial least squares discriminant analysis (PLS-DA) and random forests (RF), have been recently used to model reflectance spectral data in general, and of soil properties in particular, since their spectra are multivariate and highly collinear. Preprocessing transformations (PPTs) can improve the classification accuracy by increasing the variability between classes while decreasing the variability within classes. Such PPTs are common practice prior to a PLS-DA, but are rarely used for RF. The objectives of this paper are twofold: to compare the performances of PLS-DA and RF for modeling the spectral reflectance of soil in changed land-uses with different treatments and to compare the effects of nine different PPTs on the prediction accuracy of each of these classification methods. Differences in six physical, biological, and chemical soil properties of changed land-uses from the northern Negev Desert in Israel were evaluated. Significant differences were found between soil properties, which are used to classify land-uses and treatments. Depending on the dataset, different PPTs improved the classification accuracy by 11%–24% and 32%–42% for PLS-DA and RF, respectively, in comparison to the spectra without PPT. Out of the PPTs tested, the generalized least squares weighting (GLSW)-based transformations were found to be the most effective for most classifications using both PLS-DA and RF. Our results show that both PLS-DA and RF are suitable classifiers for spectral data, provided that an appropriate PPT is applied.
The aim of the present work was to design a portable instrument to predict the constituents of total nitrogen and organic matter in soil. Data from recent reports as well as our experiments show that these constituents exhibit absorption peaks in the near-infrared region around 850 nm and 940 nm. We designed a portable instrument that adopted diffuse reflectance spectroscopy. Subsequently, we conducted instrument performance tests to obtain the curves relating the actual and calculated values. Experimental results showed that the coefficients of determination (R2) for predicted total nitrogen and organic matter in oven-dried soil sample were 91.84% and 81.06%, respectively, and the corresponding values in wet samples were 90.17% and 80.24%. For soil organic matter and total nitrogen content in the samples, we found clear correlations between the actual values and calculated values.
Near infrared diffuse reflectance spectrophotometry, within the wavelength range 1100 to 2500 nm, was investigated for use in the simultaneous prediction of the moisture, organic C, and total N contents of air‐dried soils. An infraAlyzer 500 C (Technicon Instruments Corp.) scanning spectrophotometer was used to obtain near infrared reflectance of soils at 2‐nm intervals. Calibration equations for each of the soil constituents studied were based upon selection of the best combination of three wavelengths in a multiple regression analysis. The wavelengths selected for moisture, organic C, and total N, respectively, were 1926, 1954, and 2150 nm, 1744, 1870 and 2052 nm, and 1702, 1870 and 2052 nm. The standard errors of prediction for finely ground samples (<0.25 mm) from the top layers (0‐0.1, 0.1‐0.2, 0.2‐0.3, 0.3‐0.6 m) were 0.58, 0.16, and 0.014% for moisture, organic C, and total N, respectively. The standard errors of prediction, however, were much larger for coarsely ground soils (<2 mm), soils containing low amounts of organic C (<0.3%) and total N (<0.03%), and for those with a wide range in colors. Within a narrow range in soil color and at moderate amounts of organic matter (0.3–2.5%C), the near infrared reflectance technique provides a rapid, nondestructive, and simultaneous measurement of moisture, organic C and total N in soils
The feasibility of near infrared (NIR) reflectance spectroscopy in determining various soil constituents such as total organic carbon, total nitrogen, exchangeable potassium and available phosphorus has been investigated, to monitor their concentration during a long-term agronomic trial. Soil samples previously analysed by conventional chemical methods were scanned using a NIRSystems 5000 monochromator and spectra were treated using several algorithms. The first derivative of each NIR spectrum was used for all statistical analyses. Step-up, stepwise and modified partial least squares (MPLS) regression methods were applied to develop reliable calibration models between the NIR spectral data and the results of wet analyses. MPLS almost always gave the most successful calibrations. The results demonstrated that NIR reflectance spectroscopy can be used to determine accurately two important soil constituents, namely total nitrogen and carbon content. This technique could be employed as a routine testing method in estimating, rapidly and non-destructively, these constituents in soil samples, demonstrating soil variations within a long-term field experiment. For other determinations, such as exchangeable potassium and available phosphorus content, our results were less successful but may be useful for separation of samples into groups.
Geostatistical techniques were used to quantify the scale and degree of soil heterogeneity in 2 m 2 plots around 9-year-old poplar trees and within a wheat field. Samples were taken during two years, on an unaligned grid, for analysis of soil respiration, C and N content, available P, gravimetric moisture, pH, nitrification potential, and root biomass. Kriged maps of soil respiration, moisture, and C content showed strong spatial structure associated with poplar trees but not with wheat rows. All soil properties showed higher autocorrelation in June than in April. Isopleth patchiness for all variates was less in June. This was associated with lower respiration rates due to lower litter decomposition. From the degree and scale of heterogeneity seen in this study, we conclude that the main causes of soil heterogeneity at this scale (2 m 2) are likely to be found at micro scales controlled in part by plant root and plant residue patterns. These must be understood in the evaluation of ecosystem processes.
Development of a method to assess and monitor soil quality is critical to soil resource management and policy formation. To be useful, a method for assessing soil quality must be able to integrate many different kinds of data, allow evaluation of soil quality based on alternative uses or definitions and estimate soil quality for unsampled locations. In the present study we used one such method, based on non-parametric geostatistics. We evaluated soil quality from the integration of six soil variables measured at 220 locations in an agricultural field in southeastern Washington State. We converted the continous data values for each soil variable at each location to a binary variable indicator transform based on thresholds. We then combined indicator transformed data for individual soil variables into a single integrative indicator of soil quality termed a multiple variable indicator transform (MVIT). We observed that soil chemical variables, pools of soil resources, populations of microorgansims, and soil enzymes covaried spatially across the landscape. These ensembles of soil variables were not randomly distributed, but rather were systematically patterned. Soil quality maps calculated by kriging showed that the joint probabilities of meeting specific MVIT selection were influenced by the critical threshold values used to transform each individual soil quality variable and the MVIT selection criteria. If MVIT criteria adequately reflect soil quality then the kriging can produce maps of the probabilty of a soil being of good or poor quality.
Spatial patterns for seven soil chemical properties and textures were examined in two fields in southern Spain (Monclova and Caracol, province of Seville, Andalusia) in order to identify their spatial distribution for the implementation of a site-specific fertilization practice. Two sampling grids of 3520 and 3535 m were established in Caracol and Monclova, respectively. Fourteen and eight georeferenced soil samples per hectare were collected at two depths (0–0.1 and 0.25–0.35 m) in early November 1998 before fertilizing and planting the winter crop. Data were analyzed both statistically and geostatistically on the basis of the semivariogram. The spatial distribution model and spatial dependence level varied both between and within locations. Some of the soil properties showed lack of spatial dependence at both depths and at the chosen interval (lag h). Such was the case for clay, organic matter and NH4 at Monclova; and clay and NH4 at Caracol. Bray P and exchangeable K showed a strong patchy distribution at any field and depth. It is important to know the spatial dependence of soil parameters, as management parameters with strong spatial dependence (patchy distribution) will be more readily managed and an accurate site-specific fertilization scheme for precision farming more easily developed.
A fast and convenient soil analytical technique is needed for soil quality assessment and precision soil management. The main objective of this study was to evaluate the ability of near-infrared reflectance spectroscopy (NIRS) to predict diverse soil properties, Near-infrared reflectance spectra, obtained from a Perstrop NIR Systems 6500 scanning monochromator (Foss NIRSystems, Silver Spring, MD), and 33 chemical, physical, and biochemical properties were studied for 802 soil samples collected from four Major Land Resource Areas (MLRAs). Calibrations were based on principal component regression (PCR) using the first derivatives of optical density [log(YR)I for the 1300- to 2500-nm spectral range. Total C, total N, moisture, cation-exchange capacity (CEC), 1.5 MPa water, basal respiration rate, sand, silt, and Mehlich III extractable Ca were successfully predicted by NIRS (r(2) > 0.80). Some Mehlich III extractable metals (Fe, K, Mg, Mn) and exchangeable cations (Ca, Mg, and K), sum of exchangeable bases, exchangeable acidity, clay, potentially mineralizable N, total respiration rate, biomass C, and pH were also estimated by NIRS but with less accuracy (r(2) = 0.80 similar to0.50). The predicted results For aggregation (wt% > 2, 1, 0.5, 0.25 mm, and macroaggregation) were not reliable (r(2) = 0.46 similar to0.60). Mehlich III extractable Cu, P, and Zn, and exchangeable Na could not be predicted using the NIRS-PCR technique (r(2) < 0.50), The results indicate that NIRS can be used as a rapid analytical technique to simultaneously estimate several soil properties with acceptable accuracy in a very short time.
Efficient use of agro-chemicals is beneficial for farmers as well as for the environment. Spatial and temporal optimization of farm management will increase productivity or reduce the amount of agro-chemicals. This type of management is referred to as Precision Agriculture. Traditional management implicitly considers any field to be a homogeneous unit for management: fertilization, tillage and crop protection measures, for example, are not varied within a single field. The question for management is what to do when. Because of the variability within the field, this implies inefficient use of resources. Precision agriculture defines different management practices to be applied within single, variable fields, potentially reducing costs and limiting adverse environmental side effects. The question is not only what and when but also where. Many tools for management and analysis of spatial variable fields have been developed. In this paper, tools for managing spatial variability are demonstrated in combination with tools to optimize management in environmental and economic terms. The tools are illustrated on five case studies ranging from (1) a low technology approach using participatory mapping to derive fertilizer recommendations for resource-poor farmers in Embu, Kenya, (2) an example of backward modelling to analyze fertilizer applications and restrict nitrogen losses to the groundwater in the Wieringermeer in The Netherlands, (3) a low-tech approach of precision agriculture, developed for a banana plantation in Costa Rica to achieve higher input use efficiency and insight in spatial and temporal variation, (4) a high tech, forward modelling approach to derive fertilizer recommendations for management units in Zuidland in The Netherlands, and (5) a high-tech, backward modelling approach to detect the relative effects of several stress factors on soybean yield.
Spatio-temporal field soil and crop processes are important for site-specific farming. The objectives of this study were to spatially evaluate selected soil physical and chemical properties and their relationship to wheat (Triticum aestivum L.) yield, and to discuss stochastic approaches to help identify processes underlying yield variability in heterogeneous field sites. Modified grid sampling included 330 sites including a primary transect. Soil properties measured for the Ap, E if present, and upper B horizons at each site included pH, P, Zn, Cu, exchangeable rations, percentage base saturation, ration exchange capacity, bulk density, soil water contents at -10, -33, and -1500 kPa, texture, and humic matter content. Wheat grain and straw were hand-harvested on 1- by 2-m plots centered on each site. Soil water content on the primary transect was determined by neutron attenuation on nine dates. Field and primary transect means and semivariograms for a given soil or plant parameter were similar. The range of spatial dependence or autocorrelation of soil parameters ranged from 10 m for Ap horizon depth to 100 m for -1500 kPa water content of the Ap. Base saturation and available water storage capacity were cross-correlated with grain yield to a distance of +/-15 and 12.5 m, respectively. State-spare analysis was used to develop a grain yield model using these two variables, Spearman rank correlation of the soil water content data suggests that the temporal stability of soil water storage is less for shallow than for deeper soil layers.
The objective of this work was to investigate the usefulness of near infrared (NIR) reflectance spectroscopy in determining: (i) various constituents (total N, total C, active N, biomass and mineralisable N, and pH), (ii) parameters (soil source, depth from which sample was obtained, type of tillage used) and (iii) rate of application of NH4NO3 fertiliser) of low organic matter soils. A NIRSystems model 6250 spectrometer was used to scan soil samples (n = 179) obtained from experimental plots at two locations with three replicate plots under plow and no till practices at each location with three rates of NH4NO3 for each plot (2 x 3 x 2 x 3). For each of these, samples were taken from five depths for a total of 2 x 3 x 2 x 3 x 5 or 180 samples (one sample lost). The results demonstrated that NIR reflectance spectroscopy can be successfully used to determine some compositional parameters of low organic matter soils (particularly total C and total N). It is also apparent that for non-biological parameters (excluding soil type as reflected by source) such as the depth from which the sample was obtained, the rate of application of NH4NO3 fertiliser and the form of tillage used, that NIR reflectance spectroscopy is not very useful, unless a very limited set of samples is used (i.e. single tillage and location). For other determinations, such as pH, biomass N and active N, the results may be useful depending on the exact needs in question. Finally, from the results presented here, NIR reflectance spectroscopy was not successful in determining soil N mineralisable in 21 days.
Near-infrared reflectance analysis (NIRA) is a "sleeper" among spectroscopic techniques. David Wetzel of Kansas State University discusses the theory and practice of NIRA.
High resolution diffuse reflectance spectra (3113 spectral points) in the near infrared region (NIR) were recorded for 91 soil samples from Israel. Ten soil constituents (total iron [Fe2O3], aluminum [Al2O3]), silica [SiO2], potassium [K2O], and phosphorous [P3O2], loss on ignition residual [LOI], free iron oxides [Fed], aggregate size (1.5-2mm) fraction [F1], average aggregate size (mm) [AVGR], and sodium adsorption percentages [CNaP]) were measured by routine methods employed in soil laboratories. An empirical model to predict each property from its reflectance spectrum in the near infrared spectral region was developed by adapting the near infrared analysis (NIRA) technique. Several data manipulations were used in order to obtain optimum performance. The optimum performance of several soil constituents was found to be at 3113 spectral points (Al2O3, Fed, and K2O) and 310 spectral points (Fe2O3), whereas for others (SiO2, AVGR, and F1) even 25 spectral points provided sufficient performance. Strong support for the capability of NIRA was obtained by a careful examination of the possible correlation between spectrally active soil properties (clay content [CLAY], specific surface area [SSA], hygroscopic moisture [HIGF] and calcite [CaCO3]), which were studied elsewhere, and the featureless constituents studied here. A slight bias was found for the prediction of Al2O3 and Fed, and a greater bias was found for K2O, suggesting that further study regarding the prediction of these constituents is needed. It was concluded that NIRA is a very promising vehicle for rapid and nonrestrictive analysis of soil materials.
The objective of this work was to investigate the usefulness of near infrared (NIR) reflectance spectroscopy in determining: (i) various constituents (total N, total C, active N, biomass and mineralisable N, and pH), (ii) parameters (soil source, depth from which sample was obtained, type of tillage used) and (iii) rate of application of NH4NO3 fertiliser) of low organic matter soils. A NIRSystems model 6250 spectrometer was used to scan soil samples (n = 179) obtained from experimental plots at two locations with three replicate plots under plow and no till practices at each location with three rates of NH4NO3 for each plot (2x3x2x3). For each of these, samples were taken from five depths for a total of 2x3x2x3x5 or 180 samples (one sample lost). The results demonstrated that NIR reflectance spectroscopy can be successfully used to determine some compositional parameters of low organic matter soils (particularly total C and total N). It is also apparent that for non-biological parameters (excluding soil type as reflected by source) such as the depth from which the sample was obtained, the rate of application of NH4NO3 fertiliser and the form of tillage used, that NIR reflectance spectroscopy is not very useful, unless a very limited set of samples is used (i.e. single tillage and location). For other determinations, such as pH, biomass N and active N, the results may be useful depending on the exact needs in question. Finally, from the results presented here, NIR reflectance spectroscopy was not successful in determining soil N mineralisable in 21 days.
Near infrared (NIR) reflectance spectroscopy was used to predict the content of silt, sand, clay, iron (Fe), copper (Cu), manganese (Mn) and zinc (Zn) in soil. A total of 332 samples from agricultural soils (0-15 cm depth) in Uruguay (South America) were used. The samples were scanned in a monochromator instrument (NIRSystems 6500, Silver Spring, MD, USA). Two mathematical treatments (first and second derivative) with SNVD (scatter normal variate and detrend) and without scatter correction were studied. Modified partial least squares (mPLS) was used to develop the calibration models. The coefficient of determination in calibration (R2cal) and the standard error in calibration (SEC) using the second derivative were 0.81 (SEC: 5.1), 0.83 (SEC: 5.3), 0.92 (SEC: 2.6) for percent sand, silt and clay, respectively. The R2cal and standard error of cross-validation (SECV) were for Cu 0.87 (SEC: 0.7), for Fe 0.92 (SEC: 21.7), for Mn 0.72 (SEC: 83.0) and for Zn 0.72 (SEC: 1.2) on mg kg-1 dry matter. It was concluded that NIR reflectance spectroscopy has a great potential as an analytical method for routine analysis of soil texture, Fe, Zn and Cu due the speed and low cost of analysis.
The objective of this work was to investigate the usefulness of near infrared (NIR) reflectance spectroscopy in determining biological activity in agricultural soils. A Foss-NIRSystems model 6500 spectrometer, equipped with a spinning sample cup module, was used to scan 179 soil samples obtained from experimental plots at two locations with three replicate plots under plough and no-till practices at each location with three rates of NH4NO3 for each plot with samples taken from five depths for a total of 180 samples (one sample lost). Biological activity as measured by four enzymes (dehydrogenase, phosphatase, arylsulfatase and urease) and nitrification potential was determined by conventional methods and NIR reflectance spectroscopy. Investigations showed NIR reflectance spectroscopy to be capable of determining biological activity as reflected by the four enzymes and nitrification potential to at least some degree. With the best R2 in the range of 0.8, the results, while positive, were not as good as found previously for many other components (i.e. total C and N) in the same sample set. Efforts at simple discrimination into high, medium and low activities were not successful, and for the most part, calibrations based on subsets, such as samples from only one location, were not found to be an improvement. Correlation analysis indicated that measures of biologically-active nitrogen might be the basis for these determinations. Finally, while further research will be needed to define clearly the basis for, limitations to and usefulness of NIR reflectance spectroscopy in determining biological activity in soil samples, the results presented indicated that NIR reflectance spectroscopy might be useful for the rapid determination of such activity in cases where extreme accuracy is not required, such as spatial mapping.
Spatio-temporal field soil and crop processes are important for site-specific farming. The objectives of this study were to spatially evaluate selected soil physical and chemical properties and their relationship to wheat (Triticum aestivum L.) yield, and to discuss stochastic approaches to help identify processes underlying yield variability in heterogeneous field sites. Modified grid sampling included 330 sites including a primary transect. Soil properties measured for the Ap, E if present, and upper B horizons at each site included pH, P, Zn, Cu, exchangeable cations, percentage base saturation, cation exchange capacity, bulk density, soil water contents at -10, -33, and -1500 kPa, texture, and humic matter content. Wheat grain and straw were hand-harvested on 1- by 2-m plots centered on each site. Soil water content on the primary transect was determined by neutron attenuation on nine dates. Field and primary transect means and semivario-grams for a given soil or plant parameter were similar. The range of spatial dependence or autocorrelation of soil parameters ranged from 10 m for Ap horizon depth to 10 m for -1500 kPa water content of the Ap. Base saturation and available water storage capacity were cross-correlated with grain yield to a distance of ±15 and 12.5 m, respectively. State-space analysis was used to develop a grain yield model using these two variables. Spearman rank correlation of the soil water content data suggests that the temporal stability of soil water storage is less for shallow than for deeper soil layers.
Knowledge of ore grades and ore reserves as well as error estimation of these values, is fundamental for mining engineers
and mining geologists. Until now no appropriate scientific approach to those estimation problems has existed: geostatistics,
the principles of which are summarized in this paper, constitutes a new science leading to such an approach. The author criticizes
classical statistical methods still in use, and shows some of the main results given by geostatistics. Any ore deposit evaluation
as well as proper decision of starting mining operations should be preceded by a geostatistical investigation which may avoid
economic failures.
Simulation models and precision agriculture practices may require more detail and certainty about soil spatial variability than provided by soil surveys. This study described soil and weed spatial variability in 50-ha subareas of two sites included in the Mississippi Delta Management Systems Evaluation Areas project. Objectives were (i) to describe the spatial variability of soil properties and (ii) to determine relationships between spatially variable weed populations and soil properties. Surface soil samples were collected at nodes of 60-m square grids prior to planting cotton (Gossypium hirsutum L.) in 1996. Fieldmoist soil was analyzed for microbial activity. Air-dried soil was used to determine soil organic C, pH, and texture. Fluometuron and either clomazone, metolachlor, or norflurazon were banded over the crop row at planting. Weed counts were taken 6 wk after herbicide application. The spatial variability of soil properties and weed populations was described using geostatistics. Soil microbiological activity exhibited limited spatial dependence, but pH, organic C, and texture semivariograms were well-described with spherical models. Although short-range (<60 m) variability was often high, the range of spatial dependence typically exceeded 120 m. Total weeds were spatially dependent both years; however, weeds susceptible to control by herbicide were not. Weed densities were significantly greater (P < 0.05) in areas that had higher organic C and finer texture. Areas of low organic C and coarse soil often had no weeds. Thus, more uniform weed control might be achieved by varying preemergence herbicide application rate. Acceptable weed control might be achieved with lower herbicide application rates in certain areas.
Site-specific farming is dependent on having large amounts of reliable data. The gap between acquiring this information and using it effectively in making agricultural management decisions has widened. Collecting only the data that can be used effectively for management decisions is important. In this paper we identify the data needed to make site-specific management decisions, describe traditional and non-traditional data collection methods, discuss how these data might be used in the decision making process, and identify future needs.
The importance of biological processes for soil functioning is frequently stressed. In spite of this, the biological soil component has seldom been used to assess soil quality, while optimal and threshold values have been identified for many physical and chemical soil parameters. In this investigation, the variation in a number of microbiological, chemical and physical variables was studied on two scales. Small scales were studied with 52 samples from a single agricultural field, and large scales with a set of 26 samples from very diverse agricultural sites in Sweden. In addition, the functional structures of the two scales were studied by means of principal component analysis (PCA). The two scales had more similarities than dissimilarities regarding variable variation and functional structures. The relationships between variables were, however, more blurred in the large than in the small scales as the influences of soil structure, climate and cropping practice were larger. The possibilities to reduce the number of variables without loss of vital information about the soil system are discussed.
To obtain basic information for site-specific soil management to improve nutrient use efficiency by plants, spatial variability of soil properties was evaluated in a 50 m × 100 m paddy field. Ninety-one surface soil samples were collected after harvest to investigate the spatial variability of their chemical properties: pH, EC, total C content, total N content, C/N ratio, contents of available P, exchangeable Ca, Mg, K, and Na. Fifty samples were also collected after transplanting to investigate that of nitrogen-related properties: total C content, total N content, C/N ratio, and contents of mineralizable N and inorganic N. Geostatistical analysis was carried out to examine within-field spatial variability using semivariograms and kriged maps as well as descriptive statistics. Descriptive statistics showed that the coefficient of variation for the EC, total C content, total N content, contents of available P, exchangeable K, Na, mineralizable N, and inorganic N exceeded 10%, suggesting a relatively high variability. Geostatistical analysis indicated a high spatial dependence for all the properties except for the pH and inorganic N content. The ranges of spatial dependence were about 20 m for EC, total C content, total N content, C/N ratio, contents of exchangeable Ca, Mg, Na, and mineralizable N, and about 40–50 m for the contents of available P and exchangeable K. Based on the results of spatial dependence, kriged maps were prepared for the properties to analyze their spatial distribution in the field. The results reflected the history of soil management of the field in addition to the characteristics of the inherent soil properties. In conclusion, rational sampling interval was evaluated at 20–50 m depending on the soil properties, and the need for site-specific soil management and possibility of precision agriculture were demonstrated even in an almost flat paddy field.
Near-infrared reflectance spectroscopy (NIRS) is a rapid and nondestructive analytical technique that can be used to quantify various soil properties. The objective of this study was to evaluate the ability of NIRS to evaluate independently organic C, inorganic C, and total N content of diverse soil materials. Samples (n = 108) were prepared by mixing soil with CaCO3, humic acid, and/or compost materials. About 30% of the samples were selected randomly for the validation set, and the remaining samples were assigned to the calibration set. NIR spectra of these samples were correlated with measured values of organic C, inorganic C, total C, total N, and C:N ratios using partial least squares regression. Leave-one-out cross-validation analysis yielded r2 values between the measured and predicted soil properties higher than 0.86 for all tested properties. Similar results were obtained from analysis of the validation set (r2>0.85). The successful prediction of total N and C:N ratios for the studied samples indicates that NIRS predictions of total N for soils are based not on the strong correlation between levels of C and N but rather on an independent response to soil N. The results indicate that NIRS can be used to quantify independently and simultaneously organic C, inorganic C, and total N for soils with diverse C and N compositions.
A low cost strategy for objective and rapid selection of soil samples from a large population was evaluated. The purpose of the strategy was to retain a maximum of the original variation in important soil properties with only a small selection of samples. The evaluation was made with emphasis on clay content, soil organic matter, cation exchange capacity, and base saturation, all of which are important factors for biochemical activities in the soil and, therefore, for soil fertility. The strategy involved use of near infrared (NIR) spectroscopy combined with principal component analysis (PCA). A 2-nm interval spectrum between 1300 and 2398 nm was recorded on 146 air-dried soil samples from the most important cultivated areas in Sweden. The samples were considered mainly Cambisols and Regosols. The first derivative of each NIR spectrum was used for PCA. Twenty soils were selected by visual examination of two-dimensional score-plots from PCA. Score-plots were made from NIR data alone, from NIR data combined with pH, and from the eight significant score vectors from PCA on NIR data, combined with pH. Two criteria for selection from these plots were applied: (i) one sample from each apparent group was selected and (ii) samples evenly distributed at the periphery of the total sample population, and one in the center, were selected. In all, six selections were made. The distributions in soil properties in the selections were compared with random selection and with the original population. It was clear that NIR could help to improve the diversity in sample selections compared with random selection. In general, peripheral selections generated a higher recovery of range and a more even distribution in soil parameters than cluster selections. For clay content and cation exchange capacity, PCA on NIR data alone gave the best results, but to improve the distribution in pH and the pH-dependent base saturation, pH had to be included in PCA. To select soil samples that are distributed in all five soil parameters to the best extent possible, we propose peripheral selection from a two-dimensional PCA plot calculated from score vectors and pH data. In the present study, this method would have reduced costs about 70% compared with wet chemistry analyzes.
(C) Williams & Wilkins 1995. All Rights Reserved.
In general, agricultural management has focused on differences between fields or on the gross differences within them. Recent developments in agricultural technology, yield mapping, Global Positioning Systems and variable rate applications, have made it possible to consider managing the considerable variation in soil and other properties within fields. This system is known as precision agriculture. More precise management of fields depends on a better understanding of the factors that affect crop input decisions. This paper examines the spatial variation in crop yield, soil nutrient status and soil pH within two agricultural fields using geostatistics. The observed properties vary considerably within each field. The relation between yield and the measured soil properties appears to be weak in general. However, the range of spatial correlation for yield, shown by the variogram, is similar to that of the soil chemical properties. In addition the latter changed little over two years. This suggests that information on the scale of variation of soil chemical properties can be derived from yield maps, which can also be used as a guide to a suitable sampling interval for soil properties.
Precision agriculture, largely the application of information and communications' technologies to in-field data gathering and management, may be regarded as `best practice' for crop growth in the future because of its twin goals of maximising economic returns whilst concurrently minimising environmental impact. The practice of precision agriculture, whether it be to differentially apply fertiliser, seed, pesticide, irrigation or tillage requires detailed knowledge of the spatial and temporal variation of crop yield components, weeds, soil-borne pests and attributes of physical, chemical and biological soil fertility. However, a detailed description of fine or even coarse scale variation in soil properties has always been difficult and costly to perform. Sensing and scanning technologies are currently being developed to more efficiently and economically describe and obtain precise information on the extent and variability of soil attributes which affect crop growth and yield. Combining these technologies with vastly improved ground positioning systems allows detailed mapping of soil resource and crop yield variability which may therefore be an important input for site-specific decision making. Experiments were conducted to design an invasive sensor for real-time, simultaneous measurements of clay, organic matter and soil water content from reflectance of a suitable wavelength or combination of wavelengths in the near infra-red (NIR) portion of the electromagnetic spectrum. Soil materials were prepared with varying amounts of clay, soil water and organic matter according to a response–surface design, and the reflectance spectra measured at 2-nm intervals from 1300 nm to 2500 nm. Response–surface models were fitted to the reflectance data at specified wavelengths. Reflectance showed significant response to clay content and soil water but not to organic matter. A thorough selection procedure using non-linear modelling and root-mean-square-error of prediction was used to derive the four most suitable wavelengths (1600, 1800, 2000 and 2100 nm) for simultaneously measuring clay and soil water content. In a simulation experiment clay content was more accurately predicted than water content.
Use of precision farming technologies requires better understanding of soil variability in physical, hydraulic and chemical properties. Some of that variation is natural, some is the result of the management history of the field. So, to match agricultural inputs and practices to site-specific conditions, the factorial kriging algorithm (FKA) was used to analyze spatial variability in some soil physical, hydraulic and chemical properties (sand and silt concentrations, water contents corresponding to potentials of −10, −50, −100, −200, −1000 and −1500 kPa and organic C concentration), measured at two depths within a single field in north Italy. A linear model of coregionalization, including, (1) a nugget effect; (2) an exponential structure with an effective range of 120 m and (3) an exponential structure with an effective range of 350 m, was fitted to the experimental direct and cross-variograms of the properties of top layer. Cokriged regionalized factors, related to short and long-range variation, were then mapped to characterize soil variation across the field. Short-range soil variation was produced essentially by differences in soil texture, whereas long-range variation in organic carbon concentration resulted in dishomogeneity of soil water retention. Quite probably, the variation in organic carbon concentration was caused by the patchy discharge of liquid manure made on the field. FKA, combining pedological expert knowledge with geostatistical techniques, could be very useful to farmers so that each area within a field is managed appropriately.
Soil microorganisms mediate below- and aboveground processes, but it is difficult to monitor such organisms because of the inherent cryptic nature of the soil. Traditional `blind' sampling methods yield high sample variance. Coupled with low sample size, this results in low statistical power and thus high type II error rates. Consequently, when null hypotheses are rejected they are difficult to interpret further (either biologically insignificant or biologically significant but statistically insignificant). To help alleviate this problem and remove the `blindness' from belowground sampling we suggest researchers perform geostatistical analyses to describe the spatial distribution of the organisms/processes coupled with power analyses to assess required sample sizes. To illustrate this we intensively sampled the soil of a 3 m ×10 m plot from a southern Californian chaparral ecosystem and spatially-described a series of biological and chemical parameters. We then sampled again and stratified the data in relation to plant location and evaluated the probability of detecting a 30% increase in abundance for each variable. Overall, we found that soil organisms do not all function at similar scales, and preliminary spatial analyses help determine which organisms are suitable for study under the scales of interest. Furthermore, the results predict that required sample sizes and type II error rates will be significantly reduced for many belowground variables parameters when using a stratified sampling design. An understanding of how this spatial structure changes over time is also required to properly design stratifications and avoids bias. Thus, a priori spatial- and power-analyses can be useful tools in constructing sampling-strategies for belowground field studies.
A nondestructive method to determine total C and N concentrations in soil size fractions is desirable when a limited sample is available. Near-infrared reflectance spectroscopy (NIRS) was used to determine the total C and N concentrations in silt (50.0-2.0 μm) and coarse clay (2.0-0.2 μm), separated from 12 surface soils, by regressing the diffuse reflectance of near-infrared radiation with constituent concentrations determined using combustion techniques. The correlation coefficients (R2) of the calibration equations were 0.93 for C and 0.89 for N, and the standard errors associated with NIRS predictions were 6.2 g kg-1 soil for total C and 0.6 g kg-1 for total N. Equation development with only silt samples improved the accuracy of NIRS calibration equations. Coefficients of variation [CV = (standard error of performance ÷ mean of the combustion procedure) × 100] for validation sample sets ranged from 14 to 19%, which is within the acceptable range for determining inorganic elements in plant tissues. We conclude that NIRS can be used to predict C and N concentrations in soil size fractions.
Geostatistics is essential for environmental scientists. Weather and climate vary from place to place, soil varies at every scale at which it is examined, and even man-made attributes - such as the distribution of pollution - vary. The techniques used in geostatistics are ideally suited to the needs of environmental scientists, who use them to make the best of sparse data for prediction, and top plan future surveys when resources are limited. Geostatistical technology has advanced much in the last few years and many of these developments are being incorporated into the practitioner's repertoire. This second edition describes these techniques for environmental scientists. Topics such as stochastic simulation, sampling, data screening, spatial covariances, the variogram and its modeling, and spatial prediction by kriging are described in rich detail. At each stage the underlying theory is fully explained, and the rationale behind the choices given, allowing the reader to appreciate the assumptions and constraints involved.
Investigations have shown that near- and mid-infrared reflectance spectroscopy can accurately determine organic-C in soil. Efforts have also demonstrated that both can differentiate between organic and inorganic-C in soils, but the mid-infrared produces more accurate calibrations. Nevertheless, the greatest benefit would come with in situ determinations where factors such as particle size, sample heterogeneity and moisture can be important. While the variations in large (> 20 mesh) particle size can adversely effect calibration accuracy, efforts have demonstrated that the scanning of larger amounts of sample can overcome this, but the effects of moisture have not been fully explored. While under in situ conditions C distribution and sample heterogeneity are a problem for any analytical method, the rapid analysis possible with spectroscopic techniques will allow many more samples to be analyzed. In conclusion, near- and mid-infrared spectroscopy have great potential for providing the C values needed for C sequestration studies.
Multivariate Data Analysis—In Practice Soil map of the world, revised legend
- K H Esbensen
Esbensen, K.H., 2000. Multivariate Data Analysis—In Practice. (4th ed.) Camo ASA, Oslo. FAO, 1998. Soil map of the world, revised legend. World Soil Resources Report, vol. 60. FAO, Rome.
Exploring the spatial relations between cereal yield and soil chemical properties and the implications for sampling
- Frogbrook