Quantitative Soil Information Research Articles

Quantifying soil carbon in temperate peatlands using a mid-IR soil spectral library

Abstract. Traditional laboratory methods for acquiring soil information remain important for assessing key soil properties, soil functions and ecosystem services over space and time. Infrared spectroscopic modeling can link and massively scale up these methods for many soil characteristics in a cost-effective and timely manner. In Switzerland, only 10 % to 15 % of agricultural soils have been mapped sufficiently to serve spatial decision support systems, presenting an urgent need for rapid quantitative soil characterization. The current Swiss soil spectral library (SSL; n = 4374) in the mid-infrared range includes soil samples from the Biodiversity Monitoring Program (BDM), arranged in a regularly spaced grid across Switzerland, and temporally resolved data from the Swiss Soil Monitoring Network (NABO). Given that less than 2 % of the samples in the SSL originate from organic soils, we aimed to develop both an efficient calibration sampling scheme and accurate modeling strategy to estimate the soil carbon (SC) contents of heterogeneous samples between 0 and 2 m depth from 26 locations within two drained peatland regions (School of Agricultural, Forest and Food Sciences (HAFL) data set; n = 116). The focus was on minimizing the need for new reference analyses by efficiently mining the spectral information of the SSL. We used partial least square regressions (PLSRs), together with five repetitions of a location-grouped, 10-fold cross-validation, to predict SC ranging from 1 % to 52 % in the local HAFL data set. We compared the validation performance of different calibration schemes involving local models (1), models using the entire SSL combined with local samples (2), commonly referred to as spiking, and subsets of local and SSL samples optimized for the peatland target sites using the resampling local (RS-LOCAL) algorithm (3). Using local and RS-LOCAL calibrations with at least five local samples, we achieved similar validation results for predictions of SC up to 52 % (R2 = 0.93 to 0.97; bias = -0.07 to 1.65; root mean square error (RMSE) = 2.71 % to 3.89 % total carbon; ratio of performance to deviation (RPD) = 3.38 to 4.86; and ratio of performance to interquartile range (RPIQ) = 4.93 to 7.09). However, calibrations using RS-LOCAL only required five or 10 local samples for very accurate models (RMSE = 3.16 % and 2.71 % total carbon, respectively), while purely local calibrations required 50 samples for similarly accurate results (RMSE < 3 % total carbon). Of the three approaches, the entire SSL spiked with local samples for model calibration led to validations with the lowest performance in terms of R2, bias, RMSE, RPD and RPIQ. Hence, we show that a simple and comprehensible modeling approach, using RS-LOCAL together with a SSL, is an efficient and accurate strategy when using infrared spectroscopy. It decreases field and laboratory work, the bias of SSL spiking approaches and the uncertainty of local models. If adequately mined, the information in the SSL is sufficient to predict SC in new and independent study regions, even if the local soil characteristics are very different from the ones in the SSL. This will help to efficiently scale up the acquisition of quantitative soil information over space and time.

Hypotheses, machine learning and soil mapping

Hypotheses are of major importance in scientific research. In current applications of machine learning algorithms for soil mapping the hypotheses being tested or developed are often ambiguous or undefined. Mapping soil properties or classes, however, does not tell much about the dynamics and processes that underly soil genesis and evolution. When the interest in the soil map is for applications in a context different than soil science, such as for policy making or baseline production of quantitative soil information, the interpretation should be made in light of this application. If otherwise, we recommend soil scientists to provide hypotheses to accompany their research. The hypothesis is formulated at the beginning of the research and, in some cases, motivates data collection. Here we argue that when applying data-driven techniques such as machine learning, developing hypotheses can be a useful end point of the research. The spatial pattern predicted by the machine learning model and the correlation found among the covariates are an opportunity to develop hypotheses which are likely to require additional analyses and datasets to be tested. Systematically providing scientific hypotheses in digital soil mapping studies will enable the soil science community to build on previous work, and to increase the credibility of data-driven algorithms as a means to accelerate discovery on soil processes.

Simulation of spatial variability in crop leaf area index and yield using agroecosystem modeling and geophysics‐based quantitative soil information

AbstractAgroecosystem models that simulate crop growth as a function of weather conditions and soil characteristics are among the most promising tools for improving crop yield and achieving more sustainable agricultural production systems. This study aims at using spatially distributed crop growth simulations to investigate how field‐scale patterns in soil properties obtained using geophysical mapping affect the spatial variability of soil water content dynamics and growth of crops at the square kilometer scale. For this, a geophysics‐based soil map was intersected with land use information. Soil hydraulic parameters were calculated using pedotransfer functions. Simulations of soil water content dynamics performed with the agroecosystem model AgroC were compared with soil water content measured at two locations, resulting in RMSE of 0.032 and of 0.056 cm3 cm−3, respectively. The AgroC model was then used to simulate the growth of sugar beet (Beta vulgaris L.), silage maize (Zea mays L.), potato (Solanum tuberosum L.), winter wheat (Triticum aestivum L.), winter barley (Hordeum vulgare L.), and winter rapeseed (Brassica napus L.) in the 1‐ by 1‐km study area. It was found that the simulated leaf area index (LAI) was affected by the magnitude of simulated water stress, which was a function of both the crop type and soil characteristics. Simulated LAI was generally consistent with the observed LAI calculated from normalized difference vegetation index (LAINDVI) obtained from RapidEye satellite data. Finally, maps of simulated agricultural yield were produced for four crops, and it was found that simulated yield matched well with actual harvest data and literature values. Therefore, it was concluded that the information obtained from geophysics‐based soil mapping was valuable for practical agricultural applications.

Efficiency of ambient vibration HVSR investigations in soil engineering studies: backfill study in the Algiers (Algeria) harbor container terminal

This paper deals with the contribution of the ambient vibration horizontal-to-vertical spectral ratio (HVSR) method in soil engineering studies, particularly in backfill compactness assessment. The study is based on 60 ambient vibration recordings performed in 2015 at the container terminal of Algiers harbor, subjected a year before to a geotechnical study based on 23 boreholes and 13 cone penetration tests (CPT) for backfill improvement. To highlight the contribution of the HVSR method, the results of the geotechnical and HVSR studies are first analyzed separately and then in combination. The HVSR method provides a compactness zonation map based on peak amplitude variation. Both methods define the same pattern: a southern section where the backfill is more compact, and a northern section where the backfill is less compact. This shows that the HVSR peak amplitudes are sensitive to compactness variations, which may be sufficient for qualitative zonation. In addition, with the combination of the two methods, rough estimations of shear-wave velocity and thickness of the backfill can be retrieved. This study shows that the HVSR method can be a very useful investigative tool in soil engineering studies. When the HVSR method is deployed before any conventional technique, a geotechnical investigation campaign can be significantly optimized. Moreover the combined interpretation brings complementary quantitative soil information.

The Australian three-dimensional soil grid: Australia’s contribution to the GlobalSoilMap project

Information on the geographic variation in soil has traditionally been presented in polygon (choropleth) maps at coarse scales. Now scientists, planners, managers and politicians want quantitative information on the variation and functioning of soil at finer resolutions; they want it to plan better land use for agriculture, water supply and the mitigation of climate change land degradation and desertification. The GlobalSoilMap project aims to produce a grid of soil attributes at a fine spatial resolution (approximately 100 m), and at six depths, for the purpose. This paper describes the three-dimensional spatial modelling used to produce the Australian soil grid, which consists of Australia-wide soil attribute maps. The modelling combines historical soil data plus estimates derived from visible and infrared soil spectra. Together they provide a good coverage of data across Australia. The soil attributes so far include sand, silt and clay contents, bulk density, available water capacity, organic carbon, pH, effective cation exchange capacity, total phosphorus and total nitrogen. The data on these attributes were harmonised to six depth layers, namely 0–0.05 m, 0.05–0.15 m, 0.15–0.30 m, 0.30–0.60 m, 0.60–1.00 m and 1.00–2.00 m, and the resulting values were incorporated simultaneously in the models. The modelling itself combined the bootstrap, a decision tree with piecewise regression on environmental variables and geostatistical modelling of residuals. At each layer, values of the soil attributes were predicted at the nodes of a 3 arcsecond (approximately 90 m) grid and mapped together with their uncertainties. The assessment statistics for each attribute mapped show that the models explained between 30% and 70% of their total variation. The outcomes are illustrated with maps of sand, silt and clay contents and their uncertainties. The Australian three-dimensional soil maps fill a significant gap in the availability of quantitative soil information in Australia.

Digital Mapping of Soil Particle‐Size Fractions for Nigeria

There is a growing need for spatially continuous and quantitative soil information for environmental modeling and management, especially at the national scale. This study was aimed at predicting soil particle‐size fractions (PSF) for Nigeria using random forest model (RFM). Equal‐area quadratic splines were fitted to Nigerian legacy soil profile data to estimate PSFs at six standard soil depths (0–5, 5–15, 15–30, 30–60, 60–100, and 100–200 cm) using the GlobalSoilMap project specification. We applied an additive log‐ratio (ALR) transformation of the PSFs. There was a better prediction performance (based on 33% model validation) in the upper depth intervals than the lower depth intervals (e.g., R2 of 0.53; RMSE of 13.59 g kg−1 for clay at 0–5 cm and R2 of 0.16; RMSE of 15.60 g kg−1 at 100–200 cm). Overall, the PSFs show marked variations across the entire Nigeria region with a higher sand content compared with silt and clay contents and increasing clay content with soil depth. The variation in soil texture (ST) shows a progressive transition from a coarse texture (sand) along the fringes of northern Nigeria (e.g., upper part of Maiduguri and Sokoto), to finer texture (loam to clay loam) toward the western part of the Niger Delta region in the south. The inclusion of depth as a predictor variable significantly improved the prediction accuracy of RFM especially at lower depth intervals. These results could be used for producing soil function maps for national agricultural planning and in assessments of environmental sustainability.

Digitally mapping the information content of visible–near infrared spectra of surficial Australian soils

We can use soil mapping to gain a better understanding of the soil and how it varies in the landscape. Good quality data sets that represent the survey area are important to develop quantitative spatial models for soil mapping and to evaluate their outputs. Over the past three decades, scientists have become interested in rapid, non-destructive measurements of the soil using visible–near infrared (vis–NIR) (400–2500 nm) and mid infrared (mid-IR) (2500–25,000 nm) diffuse reflectance spectra. These spectra provide an integrative technique that measures the fundamental characteristics and composition of the soil, including colour, iron oxide, clay and carbonate mineralogy, organic matter content and composition, the amount of water present and particle size. If adequately summarised and exhaustively available over large areas, this information might be useful in situations where reliable, quantitative soil information is needed, such as agricultural, environmental and ecological modelling, or for digital soil mapping. The aims of this paper are to summarise the information content of vis–NIR spectra of Australian soils and to use a predictive spatial modelling approach to digitally map this information across Australia on a 3-arc second grid (around 90 m). We measured the spectra of 4606 surface soil samples from across Australia using a vis–NIR spectrometer. The soil information content of the spectra was summarised using a principal component analysis (PCA). We used model trees to derive statistical relationships between the scores of the PCA and 31 predictors that were readily available and we thought might best represent the factors of soil formation (climate, organisms, relief, parent material, time and the soil itself). The models were validated and subsequently used to produce digital maps of the information content of the spectra, as summarised by the PCA, with estimates of prediction error at 3-arc seconds pixel resolution. The most frequently used predictors at the continental scale were factors related to climate, parent material (and time), while at landscape and more local scales, they were factors related to relief, organisms and the soil. Finally, we use our maps for pedologic interpretations of the distribution of soils in Australia. Our results might be useful in situations requiring high-resolution, quantitative soil information e.g. in agricultural, environmental and ecologic modelling and for soil mapping and classification.

Mapping continuous depth functions of soil carbon storage and available water capacity

There is a need for accurate, quantitative soil information for natural resource planning and management. This information shapes the way decisions are made as to how soil resources are assessed and managed. This paper proposes a novel method for whole-soil profile predictions (to 1 m) across user-defined study areas where limited soil information exists. Using the Edgeroi district in north-western NSW as the test site, we combined equal-area spline depth functions with digital soil mapping techniques to predict the vertical and lateral variations of carbon storage and available water capacity (AWC) across the 1500 km 2 area. Neural network models were constructed for both soil attributes to model their relationship with a suite of environmental factors derived from a digital elevation model, radiometric data and Landsat imagery. Subsequent fits of the models resulted in an R 2 of 44% for both carbon and AWC. For validation at selected model depths, R 2 values ranged between 20 and 27% for carbon prediction (RMSE: 0.30–0.52 log (kg/m 3)) and between 8 and 29% for AWC prediction (RMSE: 0.01 m/m). Visually, reconstruction of splines at selected validation data points indicated an average fit with raw data values. In order to improve upon our model and validation results there is a need to address some of the structural and metrical uncertainties identified in this study. Nevertheless, the resulting geo-database of quantitative soil information describing its spatial and vertical variations is an example of what can be generated with this proposed methodology. We also demonstrate the functionality of this geo-database in terms of data enquiry for user-defined queries.