Modeling soil evolution is an important step towards understanding the complexity of the soil system and its interaction with the other systems. The major challenge confronted by pedologists until now is the ability to develop models capable of describing the complete complexity of the soil system. This paper presents the state of art overview of such a soil evolution model, SoilGen, its applications and limitations. In addition, the paper gives an overview of how the SoilGen model may be linked to landscape evolution models to model soilscape development. SoilGen is a mechanistic pedogenetic model in which soil forming processes such as clay migration, decalcification, carbon cycling, bioturbation, physical and chemical weathering coupled with water flow are simulated at multi-millennium time scale. The model has been calibrated and undergone extensive field testing, giving reasonable results at both pedon and landscape scales. However discrepancies between observed and simulated soil properties such as base saturation (BS), cation exchange capacity (CEC) and pH have been reported. These have been attributed partly to simplification of soil forming processes particularly in the weathering and chemical systems. There is therefore a need to extend the description of chemical and weathering systems in the SoilGen model. These extensions will not only improve model performance but will also enlarge its application range in simulating the genesis of typical features of more than half of the WRB-Reference Soil Groups. We also note here that although landscape evolution models have been successfully applied to model soil production and distribution, simplified and/or incomplete description of soil forming processes remain major limitations. We therefore add to the voices in scientific literature calling for integration of pedon and landscape scale models. In addition there is critical need for high quality chronosequence, climosequence, and toposequence profile datasets to enhance calibration and validation of soil evolution models.
Soil organic carbon (SOC) plays a major role in the global carbon budget. It can act as a source or a sink of atmospheric carbon, thereby possibly influencing the course of climate change. Improving the tools that model the spatial distributions of SOC stocks at national scales is a priority, both for monitoring changes in SOC and as an input for global carbon cycles studies. In this paper, we compare and evaluate two recent and promising modelling approaches. First, we considered several increasingly complex boosted regression trees (BRT), a convenient and efficient multiple regression model from the statistical learning field. Further, we considered a robust geostatistical approach coupled to the BRT models. Testing the different approaches was performed on the dataset from the French Soil Monitoring Network, with a consistent cross-validation procedure. We showed that when a limited number of predictors were included in the BRT model, the standalone BRT predictions were significantly improved by robust geostatistical modelling of the residuals. However, when data for several SOC drivers were included, the standalone BRT model predictions were not significantly improved by geostatistical modelling. Therefore, in this latter situation, the BRT predictions might be considered adequate without the need for geostatistical modelling, provided that i) care is exercised in model fitting and validating, and ii) the dataset does not allow for modelling of local spatial autocorrelations, as is the case for many national systematic sampling schemes.
The incorporation of remote sensing (RS) data into digital soil models has shown success to improve soil predictions. However, the effects of multiresolution imagery on modeling of biogeochemical soil properties in aquatic ecosystems are still poorly...
Digital terrain analysis (DTA) provides efficient, repeatable, and quantified metrics of landscape characteristics that are important to the Earth sciences, particularly for detailed soil mapping applications. However, DTA has not been field tested to the extent that traditional field metrics of topography have been. Human assessment of topography synthesizes multiple parameters at multiple scales to characterize a landscape, based on field experience. In order to capture the analysis scale used by field scientists, this study introduces a method for calibrating the analysis scale of DTA to field assessments. This method is used to calibrate land-surface derivatives of relative elevation, profile curvature, and slope gradient in the context of the commonly used field description of hillslope position. For a topographically diverse landscape in MI, USA, a peak in agreement between field assessment and DTA was found at field equivalent distances of 135 m for relative elevation, 63 m for profile curvature, and 9 m for slope gradient. Given the field experience of soil scientists, these calibrations of DTA metrics are likely to have stronger correlations with hillslope properties and could be used together to classify hillslope position consistently across large extents.
Regional estimates of soil carbon pool have been made using various approaches that combine soil maps with sample databases. The point soil organic carbon (SOC) densities are spatialized employing approaches like regression, spatial interpolation, polygon based summation, etc. The present work investigates a data mining based spatial imputation for spatial assessment of soil organic carbon density. The study area covers Andhra Pradesh and Karnataka states of India. Field sampling was done using stratified random sampling method with land cover/use, soil type, agro-ecological regions for defining strata. The spatial data at 1 km resolution on climate, NDVI, land cover, soil type, topography was used as input for modeling the top 30 cm Soil Organic Carbon (SOC) density. To model the SOC density, a Random Forest (RF) based model with optimal parameters and input variables has been adopted. Experiment results indicate that 500 number of trees with 5 variables at each split could explain the maximum variability of soil organic carbon density of the study area. Out of various input variables used to model SOC density, land use / cover was found to be the most significant factor that influences SOC density with a distinct importance score of 34.7 followed by NDVI with a score of 12.9. The predicted mean SOC densities range between 2.22 and 13.2 Kg m−2 and the estimated pool size of SOC in top 30 cm depth is 923 Tg for Andhra Pradesh and 1,029 Tg for Karnataka. The predicted SOC densities using this model were in good agreement with the measured observations (R = 0.86).
The value of soil is often neglected in developing countries, partially due to a lack of spatial soil data. Conventional methods of soil survey are too cumbersome and expensive to fulfil the need for soil maps in these countries. This study presents an expert knowledge based digital soil mapping (DSM) approach to provide in-time spatial soil information in developing countries. The objective of this study was to evaluate the potential of DSM soil survey methods to rapidly produce land suitability maps of a large area with acceptable accuracy. An expert knowledge approach was used, with soil surveyors creating conceptual soil distribution patterns, and populating the patterns with covariate values to create soil–landscape rules. A soil class map was created by running an inference with those rules. The map achieved an absolute validation accuracy of 80%, and 59% at a 95% confidence level. Land suitability maps were created based on the soil class map. Furthermore the data indicated that 14 or more soil observations are needed per homogeneous area to achieve acceptable results and that multiple scale covariates were useful to map different parts of the landscape.
The vegetation of the Campinaranas occurs in humid areas with hydromorphic sandy soils at the Amazon region. Thus, the determination and in situ monitoring of moisture content in Campinarana soils, besides the detection of subsurface layers are key measures for studying these soil–vegetation systems. Also, the application of ground penetrating radar (GPR) in deep sandy sedimentary sequence of Amazonia is a promising tool to enhance the knowledge on depositional and soil formation features.
Materials and methods
We studied representative soils of the Campinaranas at the National Park of Viruá, state of Roraima (Brazilian Amazonia), through the use of geophysical methods (soil moisture sensors and GPR). The study was applied in four sandy soils. Besides chemical and physical analysis of soils, soil moisture sensors were installed for monitoring during an entire hydrological year (2010/2011), and performed the calibration of sensors , coupled with imaging of the soil along transects with GPR.
Results and discussion
As a result of calibration of the soil moisture sensors we obtained a general equation with an R 2greater than 0.9. There is an influence of soil moisture content and soil temperature in the distribution of vegetation types in Campinaranas. The use of GPR identified some determinants characteristics in these soils for the differentiating the Campinaranas, represented by spodic and C horizons.
The spodic horizons in soils under Forest Campinarana provided potential errors in the determination of soil moisture, requiring calibration data for the precise use of this device. The investigation of the soil through the GPR showed interesting results, which allowed continuous visualization of the main soil horizons along transects in the phytophysiognomies of Campinaranas.
This paper describes the structure of JGrass-NewAge: a system for hydrological forecasting and modelling of water resources at the basin scale. It has been designed and implemented to emphasize the comparison of modelling solutions and reproduce hydrological modelling results in a straightforward manner. It is composed of two parts: (i) the data and result visualization system, based on the Geographic Information System uDig and (ii) the component-based modelling system, built on top of the Object Modelling System v3. Modelling components can be selected, adapted, and connected according to the needs of the modeller and then executed within the uDig spatial toolbox. Hence, the system provides an ideal and modern integration of models and GIS without invalidating existing solutions. Compared to traditional hydrological models, which are built upon monolithic code, JGrass-NewAge allows for multiple modelling solutions for the same physical process, provided they share similar input and output constraints. Modelling components are connected by means of a concise scripting language. Furthermore, the system utilizes the Pfafstetter numbering scheme to represent the digital watershed model; the adaption of this topological classification of a basin with respect to NewAge is explained in this paper. Finally, the system application for the Fort Cobb watershed and its results are presented.
A morphometric analysis of drainage networks and relief using geomorphic indices and geostatistical analyses of topographical data are useful tools for discussing the morphoevolution of a given area. Among the geomorphic indices, the Stream Length-Gradient (SL) index represents a practical tool to highlight anomalous changes in river gradients. Perturbations of SL are usually indicative of (1) differences in the resistance of outcropping lithological units to erosion, (2) sub-surface processes, such as active faulting, and (3) slope failures that directly reach the stream channels, particularly in small catchments. In this work, the SL index was calculated for the upstream sector of the Gállego River basin in the central Spanish Pyrenees to test its accuracy and sensitivity for detecting the imprints of different surface processes. A geostatistical procedure is proposed to obtain SL index maps through the interpolation and filtering of the values estimated along the drainage network. This method allows computing of SL, validation and assessing of its spatial distribution with robust statistical accuracy, and objectively defining the anomalies in SL. The anomalies in the SL map of the study area, which coincide with knickpoints and knickzones, were analyzed in detail. The results indicate (1) perturbation in the drainage network caused by differences in the resistance to erosion of outcropping lithological units and (2) hillslopes affected by large landslides, earth flows, and rock falls directly reaching the stream bed. This study indicates that the SL index has strong potential to solve geomorphological problems in different geological settings, especially in detecting the role of active, large-scale features that influence landscape evolution.
However, so far only a few attempts have been made to apply digital soil mapping techniques in tropical mountain landscapes, due to their often high heterogeneity, difficult terrain, and low accessibility.
Quantification of soil organic carbon (SOC) stocks is quite useful for accurate monitoring of C sequestration. However, there are still substantial gaps in our knowledge of SOC stocks in many parts of the world, including the Himalayas. We investigated the total SOC stocks and its spatial distribution under different land use and land cover (LULC) types in montane ecosystems of Bhutan. 186 Soil profiles were described and sampled by genetic horizons at sites located using conditioned Latin hypercube sampling. SOC concentrations at the standard depths designated for the GlobalSoilMap.Net were estimated with an equal-area spline profile function. SOC concentrations at these depth intervals were digitally mapped to a fine resolution matrix of 90 m grid using regression kriging. We found significant influence of LULC categories on SOC concentration, SOC density, SOC stocks and their spatial distributions, although this influence decreased with increasing soil depth. The estimated mean SOC density in the top 1 m were highest in fir forest soils (41.4 kg m−2) and lowest in paddy land (12.0 kg m−2). Allowing for LULC relative areas, mixed conifer forest had the highest SOC stocks in the upper meter (12.4 Mt) with orchards the lowest (0.1 Mt). The total SOC stocks for the whole study area for the 0–5, 5–15, 15–30, 30–60 and 60–100 cm depths were 2.6, 5.0, 6.5, 7.5 and 5.4 Mt, respectively. The overall SOC stock of the study area for the upper meter was approximately 27.1 Mt. The combined forests accounted for more than 77.5% of the total SOC stocks of the study area. This and the relative SOC densities indicate that the conversion of even a fraction of forests to other LULC types could lead to substantial loss of SOC stocks. This loss of SOC stock is even greater when the decrease in aboveground biomass is also taken into consideration. However, appropriate management of the agricultural lands could increase their sequestration of atmospheric CO2.
This study presents a new approach to classifying types of soil based on the probability classes of the relevant set of attributes. Two key ideas are addressed in this study: (i) the use of stochastic simulations to generate a local cumulative distribution function or extreme classes of each attribute and (ii) the use of a multidimensional scaling (MDS) technique to visualize and quantify the relative importance of each attribute in the classification process. After the simulated realizations, the weighted “distances” attributes extreme values (probability classes) of each grid node are calculated and the MDS algorithm is applied for the spatial representation of the grid nodes in a new Cartesian reference frame based on the “distances” of the probability classes of attributes. This allows the classification of soil types based on the clusters in the MDS space, after expert validation. In the second step, a sensitivity analysis of the attributes is performed with MDS: each attribute is made “neutral” one at a time, by assuming the median rather than the extreme values in each grid node before the distance evaluation, and the consequent impact on the shape and centroid displacement of the clusters (soil types) in the MDS reference frame is calculated. Hence, the spatial uncertainty of the soil type/classes and the influence of various properties are evaluated in the MDS reference frame. This method is applied to soils in a region of Brazilian in which the previous classification of soil types has been a crucial tool for precision agriculture management. Using the MDS algorithm, the selected attributes (horizon, textural gradient, colors, saturation, sand content, and clay content) were represented in a two-dimensional plot and grouped into eight clusters distinguished from each other by their characteristics. A sensitivity analysis shows that the horizon and saturation attributes had the greatest influence on determination of the clusters, i.e., the soil types.
This work aimed to evaluate whether different types of landscape structures (undulations, lynchets and undisturbed surfaces) can be discriminated by their morphometric attributes and the soil thickness. Three models based on the factorial discriminant analysis (FDA), the multinomial logistic regression (MLR) and the classification and regression trees (CART), respectively, were developed to classify different types of landscape structures. All these statistical techniques were performed using a training sample of 586 individuals over a 17 ha area located in the south-western Parisian Basin. The models developed by the CART and FDA revealed that in addition to soil thickness, the morphometric attributes slope and profile curvature significantly influence the spatial distribution of landscape structures. In addition to the variables selected by CART and FDA models, MLR model included elevation. An external validation of the classification models based on a validation sample of 148 individuals, revealed an overall well classification by CART model of 85% while those achieved with MLR and FDA models were 72% and 77%, respectively. As the predictor variables are known at all the nodes of a regular grid covering the study area; the three models developed were then used to map the landscape structures all over the 17 ha area. Resulting maps revealed a total disagreement between the three models for only 3% of the study area. For more than 50% of the study area the three models predicted a similar landscape structure. For the remaining surface, at least two of the three models predicted a similar landscape structure.