Pedometrics

Digital soil morphometrics is defined as the application of tools and techniques for measuring and quantifying soil profile attributes and deriving continuous depth functions. This paper reviews how proximal soil sensing and other tools can be used in soil profile descriptions where techniques and toolkits have not changed in the past decades. The application of such tools is compared to standard soil profile descriptions for 11 common attributes: horizons, texture, color, structure, moisture, mottles, consistence, carbonates, rock fragments, pores and roots. These attributes are extensively used in soil classification and are indicative of many soil functions. There has been progress in distinguishing soil horizons, texture and soil color, mainly using vis–NIR, GPR and electrical resistivity. There is potential for in situ digital morphometrics for all attributes of a soil profile. Smaller depth increments can be sampled and analyzed, and that gives continuous depth functions of soil properties. The combined use of in situ digital morphometrics and continuous depth functions of soil properties may enhance our pedological understanding. It will take time before the toolbox of the field pedologists will be digitally enriched, but we think that digital soil morphometrics has the potential to complement existing description and analytical methods. It may yield new insights in soil horizonation, how soils form and how they could be classified.

... decline in soil mineralogy, soil morphology and soil genesis research in comparison to a ... part of an ecosystem interacting with environmental factors generating complex patterns and processes ... consider broader issues such as soil and water quality, food security, carbon cycling ...

"Spatial Relationships between Soil Nematodes and Heavy Metals Contamination from Mines of La Union (Spain).

Heavy Metal Relationships Between Arable and Greenhouse Soils using a Geostatistical Analysis to Evaluate Spatial Variability in Short and Large Scale

Real Spatial Structures of Heavy Metal in Soils of Duero Basin (Spain): Geostatistical Approaches for Evaluating Contamination Inputs"

Geostatistical modelling has proven to be a good tool for decision making in soil nutrient management because it has the ability to map spatial heterogeneity and uncertainty. This study outlines a comparative approach to quantify the uncertainties and correlations in spatial process models as illustrated for the distribution of Zn in top soils of a semi-arid environment. The spatial correlation of Zn uncertainties is investigated by calculating the semi-variance of normalized Zn concentration residuals, for a sufficiently large dataset, as a function of distance between pairs of locations within the Zn field. Three semi-variogram models, namely: Exponential, Gaussian, and Spherical models were tested on the spatial correlation structure of the normalized residuals and further used as default input model for Sequential Gaussian Simulation. The Sequential Gaussian Simulation led to an improved description of spatial heterogeneity and uncertainty. The choice of semivariogram model is therefore critical for robust Sequential Gaussian Simulation and uncertainty analysis. A new criterion is proposed to inform the choice of default input model for sequential Gaus- sian simulation given by ratio of partial sill to square of major range. With this criterion, Exponential model performed better than both Gaussian and Spherical models. The results indicate spatial correlation in Zn field and it is suggested that the spatial correlation of Zn uncertainties should be considered to exclude inaccurate estimation for spatially dis- tributed systems when performing nutrient distribution studies.

We describe in this paper, a broad overview of spatial scale concepts and scaling procedures that are specifically relevant for digital soil mapping (DSM). Despite the recent growth and operational status of DSM, one existing and foreseeably growing issue for users of digital soil information is the inequality of spatial scales between what is required and what is actually available to adequately address soil-related questions posed from within and from outside the soil science community. In the absence of conducting new soil survey or not being able to acquire the original legacy soil information (soil point data) as a means of creating user-specified soil information products, spatial scaling provides a useful solution. Spatial scaling for DSM involves changes in map extent, grid-cell resolution, and prediction support. We review in this paper the different forms of spatial scaling, which are described in terms of changes to grid spacing and prediction support. Fine-gridding and coarse-gridding are operations where the grid spacing changes but support remains unchanged. Deconvolution and convolution are operations where the support always changes which may or may not involve changing the grid spacing. While disseveration and conflation operations occur when the support and grid size are equal and both are then changed equally and simultaneously. Some possible and existing pedometric methods are described for implementation of each scaling process, as is an extended example for performing convolution where the support changes yet the resolution remains the same.

Soil erosion is a treat to global food security. The objective of this study was to evaluate factors influencing erosion on the arable lands of the Jos Plateau; and to estimate the extent of soil erosion in the area. Universal Soil Loss Equation (USLE) model was used to evaluate soil erosion processes in the study area. This was facilitated with the aid of Geographic Information System Both for Interpolation and Geospatial analysis. Soil data from field survey was the primary source of data for analysis of soil erodibility. Topographic factor was determined from 90-meter elevation data. Rainfall erosivity was determined from rainfall data at 1 kilometer resolution. Whereas vegetation cover factor was determined from Normalized Difference Vegetation Index. Results of the study indicate that rainfall erosivity values were remarkably high and have mean values of 5117MJ.mm/ ha.h.y. Analysis of percent areal coverage indicate that the entire area had 52, 34, 7, and 7% low, moderate, high and very high topographic factors respectively. Further analysis indicate that anthropogenic factors had severely affected vegetation coverage of the Jos plateau, especially on the arable lands. Furthermore, during this research, the mean annual actual and potential soil erosion rates were estimated spatially over the Jos Plateau area. Soil erosion rates were far more than tolerable rates thereby affecting soil fertility and productivity.

Distinguishing between different sources of potentially toxic elements in soils can be difficult. This paper describes an application of principal component analysis (PCA) to distinguish between geogenic enrichment and anthropogenic pollution with Be, Cd, Co, Cr, Cu, Hg, Ni, Pb, and Zn at 14 localities in Northern and North-eastern Czech Republic. Element speciation, profile distribution, and local geology were used to facilitate interpretation of the PCA results. Of the total element contents in the topsoil and subsoil, a group of non-polluting elements, comprising Co, Cr, Cu, Ni, and Zn, was identified by PCA. There were more non-polluting elements in the subsoil than in the topsoil. The silicate-bound fraction was the most abundant in their speciation. They are likely to be mainly of geogenic origin, therefore. Beryllium also probably originated mainly from parent rocks. However, it had a closer relationship with geogenic Hg and Pb. Cadmium, Pb, and Hg showed strong topsoil enrichment. In speciation, there were larger proportions of their mobile and mobilizable fractions. This implies a significant contribution of anthropogenic pollution to their soil content. In addition, sites with some geochemical anomaly and polluted sites were indicated by the PCA. The analysis provided a concise summary of the complex information on both the generally prevailing origin of potentially toxic elements and the origin of elements at individual sampling localities.

The pedosphere is the largest terrestrial reservoir of organic carbon, yet soil-carbon variability and its representation in Earth system models is a large source of uncertainty for carbon-cycle science and climate projections. Much of this uncertainty is attributed to local and regional-scale variability, and predicting this variation can be challenging if variable selection is based solely on a priori assumptions due to the scale-dependent nature of environmental determinants. Data mining can optimize predictive modeling by allowing machine-learning algorithms to learn from and discover complex patterns in large datasets that may have otherwise gone unnoticed, thus increasing the potential for knowledge discovery. In this analysis, we identify important, regional-scale determinants for top- and subsoil-carbon stabilization in production forestland across the southeastern US. Specifically, we apply recursive feature elimination to a large suite of socio-environmental data to strategically select a parsimonious, yet highly predictive covariate set. This is achieved by recursively considering smaller and smaller covariate sets—or features—by first training the estimator on the full set to obtain feature importance. The least important features are pruned, and the procedure is recursively repeated until a desired number of covariates is identified. We show that although carbon ranges from 0.3 to 8.2 kg m−2 in the topsoil (0 to 20 cm), and from 0.4 to 17.6 kg m−2 in the subsoil (20 to 100 cm), this variability is predictably distributed with precipitation, soil moisture, nitrogen and sand content, gamma ray emissions, mean annual minimum temperature, and elevation. From our spatial predictions, we estimate that 2.6 Pg of soil carbon is currently stabilized in the upper 100 cm of production forestland, which covers 34.7 million ha in the southeastern US.

In many places in the world, soil information is difficult to obtain and can be non-existent. When no detailed map or soil observation is available in a region of interest, we have to extrapolate from other parts of the world. This paper
will discuss the Homosoil method, which assumes homology of soil-forming factors between a reference area and the region of interest. This includes: climate, physiography, and parent materials. The approach will involve seeking the smallest
taxonomic distance of the scorpan factors between the region of interest and other reference areas (with soil data) in the world.

Conventional soil maps represent a valuable source of information about soil characteristics, however they are subjective, very expensive, and time-consuming to prepare. Also, they do not include explicit information about the conceptual mental model ...

Pedotransfer functions (PTFs) have become a ‘white-hot’ topic in the area of soil science and environmental research. Most current PTF research focuses only on the development of new functions for predicting soil physical and chemical properties for different geographical areas or soil types while there are also efforts to collate and use the available PTFs. This paper reviews the brief history of the use of pedotransfer functions and discusses types of PTFs that exist. Different approaches to developing PTFs are considered and we suggest some principles for developing and using PTFs.We propose the concept of the soil inference systems (SINFERS), where pedotransfer functions are the knowledge rules for inference engines. A soil inference system takes measurements we more-or-less know with a given level of (un)certainty, and infers data that we do not know with minimal inaccuracy, by means of properly and logically conjoined pedotransfer functions. The soil inference system has a source, an organiser and a predictor. The sources of knowledge to predict soil properties are collections of pedotransfer functions and soil databases. The organiser arranges and categorises the PTFs with respect to their required inputs and the soil types from which they were generated. The inference engine is a collection of logical rules selecting the PTFs with the minimum variance. Uncertainty of the prediction can be assessed using Monte Carlo simulations. The inference system will return the predictions of soil physical and chemical properties with their uncertainties based on the information provided. Uncertainty in the prediction can be quantified in terms of the model uncertainty and input data uncertainty. In order to avoid extrapolation, a method was developed to quantify the degree of belonging of a soil sample within the training set of a PTF.With the first approach to a soil inference system, we can optimally predict various important physical and chemical properties from the information we have utilising PTFs as the knowledge rules.

Introduction
Soil survey, due to the complexity of collected information is a time loosing activity with high costs. Besides the survey, soil sample analysis are also costly. This high survey and analytical costs may be reduced by the adoption of new instruments as spectroradiometers. Materials & methods
Survey of Mugan Valley, Azerbaijan was carried out in the last years by Institute of Soil Science and Agro-chemistry of The National Academy of Science of Azerbaijan, Genesis, geography and soils mapping.
A related database was set up adopting the one freely available at the internet site of National Center for Soil Mapping, Italy (CREA-ABP). Soil observations, analyzed and geo-referenced have been digitalized. Sampling and analytical procedures were performed in accordance with national and international standards. The field observation  method was conducted to ISIS (L’Abate et al., 2013) and with the relative methodology (Costantini et al., 2007). Every genetic horizon were sampled and air dried.
In total 194 soil samples, out of 46 mini-pits have been collected and analyzed to determine pH, carbonates, texture, and Electric Conductibility with standard analytical procedures. Both wet and dry bare soil samples, and dry sieved samples were scanned with an ASD FieldSpec 3 spectroradiometer (350 to 2500 nm with a step of 1 nm) to set up a soil signatures digital library with a total of 1164 signatures.
The library was than used to:
1) perform a munsell color conversion model on colorimetry data
2) orientate the selection of significant samples to be analyze
The colorimetry model
The adopted scanning procedure was saving a 20 measure scan and repeating scanning up to 3 times. The best result was choose for splice correction using the ASD software VieSpecPro. Colorimetry data have been than exported to text files (CIE, 1964 illuminate D65: X, Y, Z, x, y,u', v'). The color conversion was performed adopting the Virtual Colour Atlas software (v. 2.1.0730) on CIE X, Y, Z data of the raw dry samples. To attribute the nearest Munsell soil colour the Atlas Sample CMC function was selected. Results have been compared with the manual attribution with Munsell soil colour charts.
The samples selection model
To perform the best reduced dataset to be further investigated on Organic Carbon and Nitrogen content, the signatures of the sieved dry samples were selected. K-mean Spectral analyses was performed both on different genetic horizons (A, B, C) and on the whole dataset.  The two analyses allowed us to determine two different datasets with, respectively 60 and 45 samples with a total of 103 selected samples. Out of the unselected samples were randomly selected 45 to use in the validation procedure of prediction. Four different calibrations were performed both on the two datasets, on their sum of 103 signatures, and on the whole datasets that included 149 samples. Prediction models were produced for Carbonates content and Electric Conductibility.
The Learned lesson
The colorimetry performed model highlighted several aspects of adopting colorimetry data: on one side the Munsell Hue determination seam to be consistent while the Munsell Value appears overestimated for the class 2 and the Chroma generally underestimated. The automatic attribution of Munsell color reduces the survey time and consequently positively impacts on the relatives costs but, on the other side, must be considered the higher uncertainty of modeled data.
The samples selection model highlighted that it is possible to reduce the calibration dataset maintaining relatively good results on prediction models.
References
Costantini, E.A.C.; Fantappiè, M.; L’Abate, G. (2007) Linee guida dei metodi di rilevamento e informatizzazione dei dati pedologici
L’Abate, G.; Allegri, G.; Barbera, R.; Bruno, R.; Fargetta, M.; Costantini, E.A.C. (2013) The ISIS webgis application for online Italian soil data consultation. EFITA conference, Turin, 24th-26th June, 2013 8 pp.
Virtual Colour Systems. VirtualAtlas Virtual Colour Atlas. Online: http://www.vcsconsulting.co.uk/VirtualAtlas.htm

approved: Jay S. Noller Conventional soil maps represent a valuable source of information about soil characteristics, however they are subjective, very expensive, and time-consuming to prepare. Also, they do not include explicit information about the conceptual mental model used in developing them nor information about their accuracy, in addition to the error associated with them. Decision tree analysis (DTA) was successfully used in retrieving the expert knowledge embedded in old soil survey data. This knowledge was efficiently used in developing predictive soil maps for the study areas in Benton and Malheur Counties, Oregon and accessing their consistency. A retrieved soil-landscape model from a reference area in Harney County was extrapolated to develop a preliminary soil map for the neighboring unmapped part of Malheur County. The developed map had a low prediction accuracy and only a few soil map units (SMUs) were predicted with significant accuracy, mostly those shallow SMUs t...

We review various recent approaches to making digital soil maps based on geographic information systems (GIS) data layers, note some commonalities and propose a generic framework for the future. We discuss the various methods that have been, or could be, used for fitting quantitative relationships between soil properties or classes and their ‘environment’. These include generalised linear models, classification and regression trees, neural networks, fuzzy systems and geostatistics. We also review the data layers that have been, or could be, used to describe the ‘environment’. Terrain attributes derived from digital elevation models, and spectral reflectance bands from satellite imagery, have been the most commonly used, but there is a large potential for new data layers. The generic framework, which we call the scorpan-SSPFe (soil spatial prediction function with spatially autocorrelated errors) method, is particularly relevant for those places where soil resource information is limited. It is based on the seven predictive scorpan factors, a generalisation of Jenny's five factors, namely: (1) s: soil, other or previously measured attributes of the soil at a point; (2) c: climate, climatic properties of the environment at a point; (3) o: organisms, including land cover and natural vegetation; (4) r: topography, including terrain attributes and classes; (5) p: parent material, including lithology; (6) a: age, the time factor; (7) n: space, spatial or geographic position. Interactions (*) between these factors are also considered. The scorpan-SSPFe method essentially involves the following steps: (i) Define soil attribute(s) of interest and decide resolution ρ and block size β.(ii) Assemble data layers to represent Q.(iii) Spatial decomposition or lagging of data layers.(iv) Sampling of assembled data (Q) to obtain sampling sites.(v) GPS field sampling and laboratory analysis to obtain soil class or property data.(vi) Fit quantitative relationships (observing Ockham's razor) with autocorrelated errors.(vii) Predict digital map.(viii) Field sampling and laboratory analysis for corroboration and quality testing.(ix) If necessary, simplify legend or decrease resolution by returning to (i) or improve map by returning to (v).Finally, possible applications, problems and improvements are discussed.

Legacy soil data form an important resource for digital soil mapping and are essential for calibration of models for predicting soil properties from environmental variables. Such data arise from traditional soil survey. Methods of soil survey are generally empirical and based on the mental development of the surveyor, correlating soil with underlying geology, landforms, vegetation and air-photo interpretation. There are no statistical criteria for traditional soil sampling, and this may lead to biases in the areas being sampled. The challenge is to test the use of legacy data for large-area mapping (e.g. national or continental extents) in order to limit the funds of field survey for large-area mapping. The problem is then to assess the reliability and quality of the legacy soil databases that have been mainly populated by traditional soil survey, and if there is a possibility of additional funding for sampling, to determine where new sampling units should be located. This additional sampling can be used to improve and validate the prediction model.Latin hypercube sampling (LHS) has been proposed as a sampling design for digital soil mapping when there is no prior sample. We use the principle of hypercube sampling to assess the quality of existing soil data and guide us to locations that need to be sampled.First an area is defined and the empirical environmental data layers or covariates are identified on a regular grid. The existing soil data are matched with the environmental variables. The HELS algorithm is used to check the occupancy of the legacy sampling units in the hypercube of the quantiles of the covarying environmental data. This is to determine whether legacy soil survey data occupy the hypercube uniformly or if there is over- or under-observation in the partitions of the hypercube. It also allows posterior estimation of the apparent probability of sample units being surveyed. From this information we can design further sampling. The methods are illustrated using legacy soil samples from Edgeroi, New South Wales, Australia, and from a large part of the Danube Basin. One third of the total number of sampling units are added to the original dataset. These new sampling units improve the representation of the feature space of the covariate. The standard deviation of the overall density is consequently smaller.

Abstract Techniques that disaggregate complex soil-terrain polygons from legacy maps are becoming more relevant, as cost effective highly detailed soil information is required to advise agriculture, hydrology, ecology, engineering, and a variety of other disciplines. Disaggregation involves the spatial placement of individual soil classes from soil legacy polygons which have multiple soil classes, while specifying the approximate proportion of each soil class and verbally or diagrammatically explaining their distribution in the landscape. One of the most common disaggregation approaches is known as DSMART (“Disaggregation and Harmonisation of Soil Map Units through Resampled Classification Trees”). However, DSMART is computationally intensive and has many parameters that must be optimised. This study aimed to address these drawbacks including input map selection, feature selection, and resample size optimisation. The research site was selected in the upper reaches of the Mvoti river catchment covering 317  k m 2 in KwaZulu Natal province, South Africa. The catchment consists of 20 soil-terrain polygons drawn at a 1:250,000 scale from the South African Land Type Survey (LTS). First, the optimal input map derived from landform elements (geomorphons) was selected through a spatially resampled Cramer’s V test to determine the association between the legacy polygons (proportion of terrain) and the geomorphon units. This was done for five different aggregated geomorphons with different parameters. Second, three feature selection algorithms (FSAs) were embedded into DSMART to determine if the algorithms could improve accuracy and computational efficiency. Third, the FSAs were compared using 25, 50, 100, and 200 resamples per polygon. The results indicate that the Cramer’s V test is a rapid method to determine the optimal input map. All FSAs achieved a significantly greater accuracy then when disaggregating the original legacy polygons and were more computationally efficient than when using all 52 covariates. This study has implications when disaggregating large and small datasets by improving computational efficiency while maintaining an acceptable accuracy.

The use of the apparent electrical conductivity (ECa) to map soil characteristics such as clay content is mostly limited to the field scale because of field-specific random effects e.g. due to different management history. Soil conductivity maps that are continuous across field boundaries are however a prerequisite for mapping soil properties at the landscape scale. Therefore we propose a geostatistical framework for the homogenization of soil data across field boundaries, i. e. for “matching” adjacent ECa surveys together. It consists of transforming field-specific data by minimizing the discontinuity error of an interpolator of the target variable on adjacent fields. We apply the homogenization technique in a case study in Central Germany, which shows that the proposed scaling technique is able to remove field-specific effects. A simulation study with known, simulated field-specific effects is performed to test the validity of the technique. It confirms that the algorithm is able to reduce discontinuity errors, yielding a homogenized ECa surface that is continuous across field boundaries. It identifies linear scaling parameters between fields apparently with a small bias. Spatial decorrelation between adjacent fields may lead to significant estimation uncertainties, depending on the distance between ECa survey traces and measurement error.