ORCID Profile
0000-0002-1182-2371
Current Organisation
University of Sydney
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
In Research Link Australia (RLA), "Research Topics" refer to ANZSRC FOR and SEO codes. These topics are either sourced from ANZSRC FOR and SEO codes listed in researchers' related grants or generated by a large language model (LLM) based on their publications.
Soil Sciences | Land Capability and Soil Degradation | Land Capability And Soil Degradation | Soil And Water Sciences Not Elsewhere Classified | Natural Resource Management | Carbon Sequestration Science | Earth Sciences Not Elsewhere Classified | Soil Physics | Surfacewater Hydrology | Natural Hazards | Soil Sciences not elsewhere classified | Environmental Science and Management | Crop and Pasture Production | Physical Geography and Environmental Geoscience | Soil Biology | Geomorphology and Regolith and Landscape Evolution | Environmental Impact Assessment | Soil Biology | Agro-ecosystem Function and Prediction | Agronomy | Agricultural Spatial Analysis and Modelling | Soil Physics |
Farmland, Arable Cropland and Permanent Cropland Soils | Environmental and resource evaluation not elsewhere classified | Forest and Woodlands Soils | Land and water management | Sparseland, Permanent Grassland and Arid Zone Soils | Land and water management | Field crops | Earth sciences | Global climate change adaptation measures | Forestry | Land and water management | Other | Climate variability | Land and Water Management of environments not elsewhere classified | Climate change | Management of Water Consumption by Plant Production | Mountain and High Country Soils | Ecosystem Assessment and Management of Urban and Industrial Environments | Urban Land Evaluation | Other environmental aspects | Rural Land Evaluation
Publisher: Elsevier BV
Date: 02-2015
Publisher: Elsevier BV
Date: 06-2017
Publisher: American Geophysical Union (AGU)
Date: 09-2012
DOI: 10.1029/2012GB004406
Publisher: Elsevier BV
Date: 12-2017
DOI: 10.1016/J.SCITOTENV.2017.07.201
Abstract: Understanding the uncertainty in spatial modelling of environmental variables is important because it provides the end-users with the reliability of the maps. Over the past decades, Bayesian statistics has been successfully used. However, the conventional simulation-based Markov Chain Monte Carlo (MCMC) approaches are often computationally intensive. In this study, the performance of a novel Bayesian inference approach called Integrated Nested Laplace Approximation with Stochastic Partial Differential Equation (INLA-SPDE) was evaluated using independent calibration and validation datasets of various skewed and non-skewed soil properties and was compared with a linear mixed model estimated by residual maximum likelihood (REML-LMM). It was found that INLA-SPDE was equivalent to REML-LMM in terms of the model performance and was similarly robust with sparse datasets (i.e. 40-60 s les). In comparison, INLA-SPDE was able to estimate the posterior marginal distributions of the model parameters without extensive simulations. It was concluded that INLA-SPDE had the potential to map the spatial distribution of environmental variables along with their posterior marginal distributions for environmental management. Some drawbacks were identified with INLA-SPDE, including artefacts of model response due to the use of triangle meshes and a longer computational time when dealing with non-Gaussian likelihood families.
Publisher: Elsevier BV
Date: 2018
Publisher: Hindawi Limited
Date: 27-10-2017
DOI: 10.1111/AJGW.12314
Publisher: Elsevier BV
Date: 05-2019
Publisher: CSIRO Publishing
Date: 2001
DOI: 10.1071/SR00065
Abstract: The different classification of particle-size fractions used in Australia compared with other countries presents a problem for the immediate adoption of the exotic pedotransfer functions. Australia adopted the international system which defined silt as particles with diameters in the range 2–20 m, while the USDA/FAO define it as 2–50 m. We present empirical equations to convert between the two systems. The USDA/FAO textural classes were also plotted in the International system’s coordinate. The USDA/FAO classes in the International system had a ‘boomerang’ shape and only occupy 60% of the triangle. Particle-size data showed that the data are evenly distributed in the USDA/FAO triangle, while most data are concentrated in the boomerang in the International system. We therefore suggest that it would seem wise for most countries to consider adopting the particle-size limits and texture classes of the USDA/FAO system.
Publisher: Elsevier BV
Date: 06-1999
Publisher: Elsevier BV
Date: 06-2021
Publisher: Wiley
Date: 05-2020
DOI: 10.1002/SAJ2.20053
Publisher: Wiley
Date: 11-2017
Publisher: Wiley
Date: 24-05-2013
DOI: 10.2136/VZJ2012.0141
Publisher: Elsevier
Date: 2016
Publisher: Wiley
Date: 09-2008
Publisher: Elsevier BV
Date: 03-2017
Publisher: MDPI AG
Date: 18-12-0012
DOI: 10.3390/RS14061459
Abstract: There was an error in the original publication [...]
Publisher: Elsevier BV
Date: 09-2015
Publisher: Springer Science and Business Media LLC
Date: 02-05-2015
Publisher: Wiley
Date: 17-06-2016
Publisher: Elsevier BV
Date: 11-2022
Publisher: Elsevier BV
Date: 12-2017
Publisher: Wiley
Date: 09-2019
DOI: 10.1111/SUM.12537
Publisher: Elsevier BV
Date: 15-02-2010
Publisher: Elsevier BV
Date: 12-2022
Publisher: Informa UK Limited
Date: 26-07-2017
Publisher: Elsevier BV
Date: 04-2016
Publisher: Elsevier BV
Date: 09-2002
Publisher: MDPI AG
Date: 16-10-2020
DOI: 10.3390/RS12203384
Abstract: It is widely acknowledged that the global stock of soil and environmental resources are diminishing and under threat. This issue stems from current and historical unsustainable management practices, leading to degraded landscapes, which is further compounded by increased pressures upon them from ever-increasing anthropogenic activities. To curb the trajectory toward a collapse of our ecosystems, systematic ways are needed to assess the condition of our natural resources, how much they might have changed, and to what extent this might impact on the life sustaining functions we derive from our environment and the extent of our food producing systems. Some solutions to these issues come in the form of measurement, mapping and monitoring technology, which facilitates powerful ways in which to be informed about and to understand and assess the condition of our landscapes so that they can be managed strategically or simply improved. This Special Issue showcases from several locations across the globe, detailed ex les of what is achievable at the convergence of big data brought about by remote and proximal sensing platforms, advanced statistical modelling and computing infrastructure to understand and monitor our ecosystems better. These utilities not only provide high-resolution abilities to map the extent and changes to our food producing systems, they also have yielded new ways to determine land-use and climate effects on the fate of soil carbon across living generations and to identify hydrological risk strategies in otherwise data-poor urban environments. Leveraging the availability of remote sensing data is telling, but the papers in this Special Issue also highlight the sophistication of modelling capabilities to deliver not only highly detailed maps of temporal dynamic soil phenomena but ways to draw new inferences from sparse and disparate model input data. The challenges of restoring our ecosystems are immense and sobering. However, we are well equipped and capable of confronting these pervasive issues in objective and data-informed ways that have previously never been possible.
Publisher: Elsevier BV
Date: 11-2020
Publisher: Elsevier BV
Date: 04-2000
Publisher: Wiley
Date: 05-2008
Publisher: Wiley
Date: 30-04-2013
Publisher: Elsevier BV
Date: 09-2017
Publisher: Wiley
Date: 10-2010
Publisher: CSIRO Publishing
Date: 2006
DOI: 10.1071/SR06060
Abstract: Vineyard soil surveys to date have focused on presenting soil data in point rather than raster format. This is due to the recording of both numeric and categorical variables. A protocol, including a lookup table to transform linguistic texture values into particle size distributions, to convert point data into continuous raster maps is presented. The resulting maps are coherent with vineyard knowledge and provide a strong spatial representation of soil variability within the vineyard. Validation with an independent dataset shows an error of ~10% in prediction however, some of this can be attributed to errors in the geo-rectification of old data. Raster maps allow the survey data to be incorporated into computer systems to better model vineyard and irrigation designs and are more readily used in day-to-day vineyard management decisions.
Publisher: Elsevier BV
Date: 11-2006
Publisher: Elsevier BV
Date: 07-2018
Publisher: Elsevier BV
Date: 08-1999
Publisher: Elsevier BV
Date: 07-2018
Publisher: PeerJ
Date: 16-04-2018
DOI: 10.7717/PEERJ.4659
Abstract: Soil colour is often used as a general purpose indicator of internal soil drainage. In this study we developed a necessarily simple model of soil drainage which combines the tacit knowledge of the soil surveyor with observed matrix soil colour descriptions. From built up knowledge of the soils in our Lower Hunter Valley, New South Wales study area, the sequence of well-draining → imperfectly draining → poorly draining soils generally follows the colour sequence of red → brown → yellow → grey → black soil matrix colours. For each soil profile, soil drainage is estimated somewhere on a continuous index of between 5 (very well drained) and 1 (very poorly drained) based on the proximity or similarity to reference soil colours of the soil drainage colour sequence. The estimation of drainage index at each profile incorporates the whole-profile descriptions of soil colour where necessary, and is weighted such that observation of soil colour at depth and/or dominantly observed horizons are given more preference than observations near the soil surface. The soil drainage index, by definition disregards surficial soil horizons and consolidated and semi-consolidated parent materials. With the view to understanding the spatial distribution of soil drainage we digitally mapped the index across our study area. Spatial inference of the drainage index was made using Cubist regression tree model combined with residual kriging. Environmental covariates for deterministic inference were principally terrain variables derived from a digital elevation model. Pearson’s correlation coefficients indicated the variables most strongly correlated with soil drainage were topographic wetness index (−0.34), mid-slope position (−0.29), multi-resolution valley bottom flatness index (−0.29) and vertical distance to channel network (VDCN) (0.26). From the regression tree modelling, two linear models of soil drainage were derived. The partitioning of models was based upon threshold criteria of VDCN. Validation of the regression kriging model using a withheld dataset resulted in a root mean square error of 0.90 soil drainage index units. Concordance between observations and predictions was 0.49. Given the scale of mapping, and inherent subjectivity of soil colour description, these results are acceptable. Furthermore, the spatial distribution of soil drainage predicted in our study area is attuned with our mental model developed over successive field surveys. Our approach, while exclusively calibrated for the conditions observed in our study area, can be generalised once the unique soil colour and soil drainage relationship is expertly defined for an area or region in question. With such rules established, the quantitative components of the method would remain unchanged.
Publisher: MDPI AG
Date: 13-04-2022
DOI: 10.3390/RS14081875
Abstract: Rice is the staple crop for more than half the world’s population, but there is a lack of high-resolution maps outlining rice areas and their growth stages. Most remote sensing studies map the rice extent however, in tropical regions, rice is grown throughout the year with variable planting dates and cropping frequency. Thus, mapping rice growth stages is more useful than mapping only the extent. This study addressed this challenge by developing a phenology-based method. The hypothesis was that the unsupervised classification (k-means clustering) of Sentinel-1 and 2 time-series data could identify rice fields and growth stages, because (1) the presence of flooding during transplanting can be identified by Sentinel-1 VH backscatter and (2) changes in the canopy of rice fields during growth stages (vegetative, generative, and ripening phases) up to the point of harvesting can be identified by Normalized Difference Vegetation Index (NDVI) time series. Using the proposed method, this study mapped rice field extent and cropping calendars across Peninsular Malaysia (131,598 km2) on the Google Earth Engine (GEE) platform. The Sentinel-1 and 2 monthly time series data from January 2019 to December 2020 were classified using k-means clustering to identify areas with similar phenological patterns. This approach resulted in 10-meter resolution maps of rice field extent, intensity, and cropping calendars. Validation using very high-resolution street view images from Google Earth showed that the predicted map had an overall accuracy of 95.95%, with a kappa coefficient of 0.92. In addition, the predicted crop calendars agreed well with the local government’s granary data. The results show that the proposed phenology-based method is cost-effective and can accurately map rice fields and growth stages over large areas. The information will be helpful in measuring the achievement of self-sufficiency in rice production and estimates of methane emissions from rice cultivation.
Publisher: Elsevier BV
Date: 04-2023
Publisher: Informa UK Limited
Date: 28-01-2014
Publisher: Elsevier BV
Date: 03-2021
Publisher: Elsevier BV
Date: 2021
Publisher: Elsevier BV
Date: 03-2021
Publisher: Elsevier BV
Date: 06-2016
Publisher: Elsevier BV
Date: 12-2022
Publisher: Elsevier BV
Date: 05-2017
Publisher: Elsevier BV
Date: 02-2018
Publisher: Elsevier BV
Date: 2018
Publisher: Elsevier BV
Date: 05-2022
Publisher: Elsevier BV
Date: 12-2009
Publisher: Elsevier BV
Date: 06-2021
Publisher: Wiley
Date: 06-10-2017
DOI: 10.1111/EJSS.12475
Abstract: Soil water‐holding capacity is an important component of the water and energy balances of the terrestrial biosphere. It controls the rate of evapotranspiration, and is a key to crop production. It is widely accepted that the available water capacity in soil can be improved by increasing organic matter content. However, the increase in amount of water that is available to plants with an increase in organic matter is still uncertain and may be overestimated. To clarify this issue, we carried out a meta‐analysis from 60 published studies and analysed large databases (more than 50 000 measurements globally) to seek relations between organic carbon (OC) and water content at saturation, field capacity, wilting point and available water capacity. We show that the increase in organic carbon in soil has a small effect on soil water content. A 1% mass increase in soil OC (or 10 g C kg −1 soil mineral), on average, increases water content at saturation, field capacity, wilting point and available water capacity by: 2.95, 1.61, 0.17 and 1.16 mm H 2 O 100 mm soil −1 , respectively. The increase is larger in sandy soils, followed by loams and is least in clays. Overall the increase in available water capacity is very small 75% of the studies reported had values between 0.7 and 2 mm 100 mm −1 with an increase of 10 g C kg −1 soil. Compared with reported annual rates of carbon sequestration after the adoption of conservation agricultural systems, the effect on soil available water is negligible. Thus, arguments for sequestering carbon to increase water storage are questionable. Conversely, global warming may cause losses in soil carbon, but the effects on soil water storage and its consequent impact on hydrological cycling might be less than thought previously. We investigated how available water capacity can be increased with a 1% increase in soil organic carbon. We analysed data from 60 published studies and global databases with more than 50 000 measurements. The increase in organic carbon in soil has a small effect on soil water retention. A 1% mass increase in soil OC on average increased available water capacity by 1.16%, volumetrically.
Publisher: Wiley
Date: 11-10-2007
Publisher: Elsevier BV
Date: 06-2021
Publisher: Elsevier BV
Date: 12-2017
Publisher: Elsevier BV
Date: 04-2012
Publisher: Soil and Water Conservation Society
Date: 2018
Publisher: Elsevier BV
Date: 04-2019
Publisher: Elsevier BV
Date: 03-2017
Publisher: Elsevier BV
Date: 11-2021
Publisher: Elsevier BV
Date: 02-2021
Publisher: American Geophysical Union (AGU)
Date: 05-2023
DOI: 10.1029/2022GB007679
Abstract: Soil Organic Carbon (SOC) turnover τ in wetlands and the corresponding governing processes are still poorly represented in numerical models. τ is a proxy to the carbon storage potential in each SOC pool and C fluxes within the whole ecosystem however, it has not been comprehensively quantified in wetlands globally. Here, we quantify the turnover time τ of various SOC pools and the governing biotic and abiotic processes in global wetlands using a comprehensively tested process‐based biogeochemical model. Globally, we found that τ ranges between 1 and 1,000 years and is controlled by anaerobic (in 78% of global wetlands area) and aerobic (15%) respiration, and by abiotic destabilization from soil minerals (5%). τ in the remaining 2% of wetlands is controlled by denitrification, sulfur reduction, and leaching below the subsoil. τ can vary by up to one order of magnitude in temperate, continental, and polar regions due to seasonal temperature and can shift from being aerobically controlled to anaerobically controlled. Our findings of seasonal variability in SOC turnover suggest that wetlands are susceptible to climate‐induced shifts in seasonality, thus requiring better accounting of seasonal fluctuations at geographic scales to estimate C exchanges between land and atmosphere.
Publisher: CSIRO Publishing
Date: 2003
DOI: 10.1071/SR02154
Abstract: This paper describes the hydraulic, structural and fundamental soil properties for 23 Vertosol horizons from 18 s ling sites in New South Wales and southern Queensland. At each site a combination of infiltration measurements and soil s ling was conducted. S les were collected for determination of the soil water characteristic, shrink–swell relationships, and fundamental soil properties such as particle size distributions, pH, electrical conductivity (EC), exchangeable cations (Ca2+, Mg2+, K+, Na+), extractable P contents, extractable sulfate and Fe contents, and CaCO3 and total C contents. Large cores were s led, impregnated with resin, and sectioned for image analysis. The program SOLICON v2.1 was used to calculate structural form parameters from the images. Measured hydraulic conductivities of the surface soils were large compared with earlier reported research for Vertosols. However, a sharp decrease in hydraulic conductivity occurred with depth in the profiles, which is assumed to be due to increased bulk densities and exchangeable sodium percentages (ESP). The data also indicated a general north–south trend in the structural development of these Vertosols. Surface soils from the northern areas, such as the Gwydir and Namoi valleys, exhibited more porous structural forms, and as a result, greater average hydraulic conductivities. This appears to be due to differences in ESP, clay content and the mineralogical suite of the clay surface s les with smaller ESPs and larger proportions of smectitic clay tended to have the greatest values of hydraulic conductivity. Other fundamental soil properties such as extractable Fe and P contents, and CaCO3 content, were found to have little or no correlation to the hydraulic or structural properties of these Vertosols, while differences in measured shrink-swell and water retention properties were largely a function of soil depth. The database developed has given an overview of the hydraulic properties of Vertosols used for cotton production in south-eastern Australia.
Publisher: Elsevier BV
Date: 08-2006
Publisher: Wiley
Date: 05-2011
Publisher: Wiley
Date: 2023
DOI: 10.1111/EJSS.13337
Abstract: Microorganisms play pivotal roles in soil processes. Metabolically related microorganisms constitute functional groups, and erse microbial functional groups control nutrient cycling in soils. This study explored environmental (i.e., rainfall, temperature) and soil factors driving the distribution of bacterial functional groups involved in soil carbon (C) cycling in paired natural and agricultural ecosystems. Soil s les were collected at a regional scale covering gradients of temperature and rainfall across two orthogonal transects (~1000 km) in New South Wales, Australia. Putative functions of bacteria were linked to two soil C fractions: particulate organic carbon (POC) and mineral‐associated organic carbon (MAOC). We found: (i) temperature and rainfall were important drivers of bacterial functional groups, while soil properties, such as pH, soil C and nitrogen (N), also presented significant contributions (ii) community structure of bacteria involved in C cycling was mainly related to POC content but not to MAOC (iii) paired s ling showed that agricultural practices had significant impacts on the composition and responses of soil bacterial functional groups. This study demonstrated the environmental regulation (e.g., temperature and rainfall) of soil microbial functional groups at large scales, which was altered by agricultural practices. Soil bacteria involved in C cycling were investigated across two ~1000 km transects. Temperature and rainfall were important drivers of bacterial functional groups at large scale. Paired s ling showed that agriculture led to a significant shift in bacterial functional groups. Community structure of bacterial functional groups were correlated with soil POC but not MAOC.
Publisher: Elsevier BV
Date: 09-2001
Publisher: Wiley
Date: 12-06-2015
Publisher: Elsevier BV
Date: 05-2019
Publisher: MDPI AG
Date: 28-05-2019
DOI: 10.3390/SOILSYSTEMS3020037
Abstract: Digital soil maps can be used to depict the ability of soil to fulfill certain functions. Digital maps offer reliable information that can be used in spatial planning programs. Several broad types of data mining approaches through Digital Soil Mapping (DSM) have been tested. The usual approach is to select a model that produces the best validation statistics. However, instead of choosing the best model, it is possible to combine all models realizing their strengths and weaknesses. We applied seven different techniques for the prediction of soil classes based on 194 sites located in Isfahan region. The mapping exercise aims to produce a soil class map that can be used for better understanding and management of soil resources. The models used in this study include Multinomial Logistic Regression (MnLR), Artificial Neural Networks (ANN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Bayesian Networks (BN), and Sparse Multinomial Logistic Regression (SMnLR). Two ensemble models based on majority votes (Ensemble.1) and MnLR (Ensemble.2) were implemented for integrating the optimal aspects of the in idual techniques. The overall accuracy (OA), Cohen's kappa coefficient index (κ) and the area under the curve (AUC) were calculated based on 10-fold-cross validation with 100 repeats at four soil taxonomic levels. The Ensemble.2 model was able to achieve larger OA, κ coefficient and AUC compared to the best performing in idual model (i.e., RF). Results of the ensemble model showed a decreasing trend in OA from Order (0.90) to Subgroup (0.53). This was also the case for the κ statistic, which was the largest for the Order (0.66) and smallest for the Subgroup (0.43). Same decrease was observed for AUC from Order (0.81) to Subgroup (0.67). The improvement in κ was substantial (43 to 60%) at all soil taxonomic levels, except the Order level. We conclude that the application of the ensemble model using the MnLR was optimal, as it provided a highly accurate prediction for all soil taxonomic levels over and above the in idual models. It also used information from all models, and thus this method can be recommended for improved soil class modelling. Soil maps created by this DSM approach showed soils that are prone to degradation and need to be carefully managed and conserved to avoid further land degradation.
Publisher: MDPI AG
Date: 11-07-2023
DOI: 10.3390/RS15143492
Abstract: Remote sensing approaches are often used to monitor land cover change. However, the small physical size (about 1–2 hectare area) of smallholder orchards and the cultivation of cocoa (Theobroma cocoa L.) under shade trees make the use of many popular satellite sensors inefficient to distinguish cocoa orchards from forest areas. Nevertheless, high-resolution satellite imagery combined with novel signal extraction methods facilitates the differentiation of coconut palms (Cocos nucifera L.) from forests. Cocoa grows well under established coconut shade, and underplanting provides a viable opportunity to intensify production and meet demand and government targets. In this study, we combined grey-level co-occurrence matrix (GLCM) textural features and vegetation indices from Sentinel datasets to evaluate the sustainability of cocoa expansion given land suitability for agriculture and soil capability classes. Additionally, it sheds light on underexploited areas with agricultural potential. The mapping of areas where cocoa smallholder orchards already exist or can be grown involved three main components. Firstly, the use of the fine-resolution C-band synthetic aperture radar and multispectral instruments from Sentinel-1 and Sentinel-2 satellites, respectively. Secondly, the processing of imagery (Sentinel-1 and Sentinel-2) for feature extraction using 22 variables. Lastly, fitting a random forest (RF) model to detect and distinguish potential cocoa orchards from non-cocoa areas. The RF classification scheme differentiated cocoa (for consistency, the coconut–cocoa areas in this manuscript will be referred to as cocoa regions or orchards) and non-cocoa regions with 97 percent overall accuracy and over 90 percent producer’s and user’s accuracies for the cocoa regions when trained on a combination of spectral indices and GLCM textural feature sets. The top five variables that contributed the most to the model were the red band (B4), red edge curve index (RECI), blue band (B2), near-infrared (NIR) entropy, and enhanced vegetation index (EVI), indicating the importance of vegetation indices and entropy values. By comparing the classified map created in this study with the soil and land capability legacy information of Bougainville, we observed that potential cocoa regions are already rated as highly suitable. This implies that cocoa expansion has reached one of many intersecting limits, including land suitability, political, social, economic, educational, health, labour, and infrastructure. Understanding how these interactions limit cocoa productivity at present will inform further sustainable growth. The tool provides inexpensive and rapid monitoring of land use, suitable for a sustainable planning framework that supports responsible agricultural land use management. The study developed a heuristic tool for monitoring land cover changes for cocoa production, informing sustainable development that balances the needs and aspirations of the government and farming communities with the protection of the environment.
Publisher: Wiley
Date: 03-2010
Publisher: Elsevier BV
Date: 08-2020
Publisher: Springer Science and Business Media LLC
Date: 03-08-2007
Publisher: CSIRO Publishing
Date: 2007
DOI: 10.1071/SR07051
Abstract: This paper aims to establish the means and ranges of clay, silt, and sand contents from field texture classes, and to investigate the differences in the field texture classes and texture determined from particle-size analysis. The results of this paper have 2 practical applications: (1) to estimate the particle size distribution and its uncertainty from field texture as input to pedotransfer functions, and (2) to examine the criteria of texture contrast soils in the Australian Soil Classification system. Estimates of clay, silt, and sand content for each field texture class are given and this allows the field texture classes to be plotted in the texture triangle. There are considerable differences between field texture classes and particle-size classes. Based on the uncertainties in determining the clay content from field texture, we establish the probability of the occurrence of a texture contrast soil according to the Australian Soil Classification system, given the texture of the B2 horizon and its overlying A horizon. I enjoy doing the soil-texture feel test with my fingers or kneading a clay soil, which is a short step from ceramics or sculpture. Hans Jenny (1984)
Publisher: PeerJ
Date: 11-03-2021
DOI: 10.7717/PEERJ.11042
Abstract: The development of portable near-infrared spectroscopy (NIRS) combined with smartphone cloud-based chemometrics has increased the power of these devices to provide real-time in-situ crop nutrient analysis. This capability provides the opportunity to address nutrient deficiencies early to optimise yield. The agriculture sector currently relies on results delivered via laboratory analysis. This involves the collection and preparation of leaf or soil s les during the growing season that are time-consuming and costly. This delays farmers from addressing deficiencies by several weeks which impacts yield potential hence, requires a faster solution. This study evaluated the feasibility of using NIRS in estimating different macro- and micronutrients in cotton leaf tissues, assessing the accuracy of a portable handheld NIR spectrometer (wavelength range of 1,350–2,500 nm). This study first evaluated the ability of NIRS to predict leaf nutrient levels using dried and ground cotton leaf s les. The results showed the high accuracy of NIRS in predicting essential macronutrients (0.76 ≤ R 2 ≤ 0.98 for N, P, K, Ca, Mg and S) and most micronutrients (0.64 ≤ R 2 ≤ 0.81 for Fe, Mn, Cu, Mo, B, Cl and Na). The results showed that the handheld NIR spectrometer is a practical option to accurately measure leaf nutrient concentrations. This research then assessed the possibility of applying NIRS on fresh leaves for potential in-field applications. NIRS was more accurate in estimating cotton leaf nutrients when applied on dried and ground leaf s les. However, the application of NIRS on fresh leaves was still quite accurate. Using fresh leaves, the prediction accuracy was reduced by 19% for macronutrients and 11% for micronutrients, compared to dried and ground s les. This study provides further evidence on the efficacy of using NIRS for field estimations of cotton nutrients in combination with a nutrient decision support tool, with an accuracy of 87.3% for macronutrients and 86.6% for micronutrients. This application would allow farmers to manage nutrients proactively to avoid yield penalties or environmental impacts.
Publisher: Elsevier BV
Date: 02-2007
Publisher: Elsevier BV
Date: 03-2019
Publisher: Elsevier BV
Date: 2016
Publisher: Copernicus GmbH
Date: 06-08-2020
Abstract: Abstract. Most soil management activities are implemented at farm scale, yet digital soil maps are commonly available at regional or national scale. Disaggregating these regional and/or national maps is applicable for farm-scale tasks, particularly in data-poor or limited situations. Although disaggregation is a frequently discussed topic in recent digital soil mapping literature, the uncertainty of the disaggregation process is not often discussed. Underestimation of inferential or predictive uncertainty in statistical modelling leads to inaccurate statistical summaries and overconfident decisions. The use of Bayesian inference allows for quantifying the uncertainty associated with the disaggregation process. In this study, a framework of Bayesian area-to-point regression kriging (ATPRK) is proposed for downscaling soil attributes, in particular, maps of soil organic carbon. An estimation of point support variograms from block-supported data was carried out using the Monte Carlo integration via the Metropolis–Hastings algorithm. A regional soil carbon map with a resolution of 100 m (block support) was disaggregated to 10 m (point support) information for a farm in northern New South Wales (NSW), Australia. The derived point support variogram has a higher partial sill and nugget, while the range and parameters do not deviate much from the block support data. The disaggregated fine-scale map (point support with a grid spacing of 10 m) using Bayesian ATPRK had an 87 % concordance correlation with the original coarse-scale map. The uncertainty estimates of the disaggregation process were given by a 95 % confidence interval (CI) limit. Narrow CI limits indicate that the disaggregation process gives a fair approximation of the mean soil organic carbon (SOC) content of the study site. The Bayesian ATPRK approach was compared with dissever, which is a regression-based disaggregation algorithm. The disaggregated maps generated by dissever had 96 % concordance correlation with the coarse-scale map. Dissever achieves this higher concordance correlation through an iteration process, while Bayesian ATPRK is a one-step process. The two disaggregated products were validated with 127 independent topsoil carbon observations. The validation concordance correlation coefficient for Bayesian ATPRK disaggregation was 23 %, while downscaled maps generated from dissever had 18 % concordance correlation coefficient (CCC). The advantages and limitations of both disaggregation algorithms are discussed.
Publisher: Elsevier BV
Date: 07-2021
Publisher: Elsevier BV
Date: 10-2009
Publisher: Copernicus GmbH
Date: 14-08-2020
Abstract: Abstract. Enhancing the spatial resolution of pedological information is a great challenge in the field of digital soil mapping (DSM). Several techniques have emerged to disaggregate conventional soil maps initially and are available at a coarser spatial resolution than required for solving environmental and agricultural issues. At the regional level, polygon maps represent soil cover as a tessellation of polygons defining soil map units (SMUs), where each SMU can include one or several soil type units (STUs) with given proportions derived from expert knowledge. Such polygon maps can be disaggregated at a finer spatial resolution by machine-learning algorithms, using the Disaggregation and Harmonisation of Soil Map Units Through Res led Classification Trees (DSMART) algorithm. This study aimed to compare three approaches of the spatial disaggregation of legacy soil maps based on DSMART decision trees to test the hypothesis that the disaggregation of soil landscape distribution rules may improve the accuracy of the resulting soil maps. Overall, two modified DSMART algorithms (DSMART with extra soil profiles DSMART with soil landscape relationships) and the original DSMART algorithm were tested. The quality of disaggregated soil maps at a 50 m resolution was assessed over a large study area (6775 km2) using an external validation based on 135 independent soil profiles selected by probability s ling, 755 legacy soil profiles and existing detailed 1:25 000 soil maps. Pairwise comparisons were also performed, using the Shannon entropy measure, to spatially locate the differences between disaggregated maps. The main results show that adding soil landscape relationships to the disaggregation process enhances the performance of the prediction of soil type distribution. Considering the three most probable STUs and using 135 independent soil profiles, the overall accuracy measures (the percentage of soil profiles where predictions meet observations) are 19.8 % for DSMART with expert rules against 18.1 % for the original DSMART and 16.9 % for DSMART with extra soil profiles. These measures were almost 2 times higher when validated using 3×3 windows. They achieved 28.5 % for DSMART with soil landscape relationships and 25.3 % and 21 % for original DSMART and DSMART with extra soil observations, respectively. In general, adding soil landscape relationships and extra soil observations constraints allow the model to predict a specific STU that can occur in specific environmental conditions. Thus, including global soil landscape expert rules in the DSMART algorithm is crucial for obtaining consistent soil maps with a clear internal disaggregation of SMUs across the landscape.
Publisher: Elsevier BV
Date: 04-2019
Publisher: Elsevier BV
Date: 06-2005
Publisher: Wiley
Date: 28-01-2013
DOI: 10.1111/EJSS.12012
Publisher: Elsevier BV
Date: 2023
DOI: 10.1016/J.SCITOTENV.2022.159253
Abstract: Increased soil organic carbon (OC) in China has been reported in the past two decades, suggesting the sequestration of atmospheric carbon dioxide into soil, mitigating climate change and improving soil health. On the other hand, soil pH decrease had also been reported nationwide. If the two are related, the strategy of increasing soil OC could negatively affect soil quality for food production and the environment. We investigate this thread based on large-scale soil survey data from two provinces with typical soil and cropping patterns in the east and south of China, Jiangsu (102,600 km
Publisher: Elsevier
Date: 2023
Publisher: Elsevier BV
Date: 12-2019
Publisher: Elsevier BV
Date: 08-2014
Publisher: CSIRO Publishing
Date: 2006
DOI: 10.1071/SR05152
Abstract: Using a range of earlier published results and a recently published dataset, pedotransfer functions (PTFs) were developed to predict some hydraulic properties of Vertosols. A fitting approach using neural networks was employed with good results to predict the soil water characteristic curve. The developed functions are complex due to the large numbers of parameters, but moisture contents are predicted to within 5%. Other PTFs to predict the moisture content at the drained upper limit (DUL) and lower limit (LL), and bulk density in the normal shrinkage curve, were developed using multiple linear regression. The PTFs to predict the soil water characteristic curve, DUL and LL, and the bulk density in the normal shrinkage zone were mainly based on total clay, sand, and silt contents and bulk density, with minor contributions of ECEC and total carbon content. PTFs for unsaturated hydraulic conductivities were also developed using multi-linear regression and were mainly dependent on silt contents and ESP values. The mean error in these predictions was 2.76 mm/h, which is reasonable for predictions at the field and farm scale where inherent soil variability can cause larger variation. The developed PTFs can be used to predict parameters needed in crop modelling tools such as OZCOT to simulate cotton development on Vertosols. Some further ex les of the use of the PTFs for management of irrigation are given.
Publisher: Elsevier BV
Date: 12-2017
Publisher: Springer Science and Business Media LLC
Date: 31-03-2023
DOI: 10.1038/S41597-023-02056-8
Abstract: We introduce a new dataset of high-resolution gridded total soil organic carbon content data produced at 30 m × 30 m and 90 m × 90 m resolutions across Australia. For each product resolution, the dataset consists of six maps of soil organic carbon content along with an estimate of the uncertainty represented by the 90% prediction interval. Soil organic carbon maps were produced up to a depth of 200 cm, for six intervals: 0–5 cm, 5–15 cm, 15–30 cm, 30–60 cm, 60–100 cm and 100–200 cm. The maps were obtained through interpolation of 90,025 depth-harmonized organic carbon measurements using quantile regression forest and a large set of environmental covariates. Validation with 10-fold cross-validation showed that all six maps had relatively small errors and that prediction uncertainty was adequately estimated. The soil carbon maps provide a new baseline from which change in future carbon stocks can be monitored and the influence of climate change, land management, and greenhouse gas offset can be assessed.
Publisher: American Geophysical Union (AGU)
Date: 11-2015
DOI: 10.1002/2015WR017703
Publisher: Wiley
Date: 24-01-2012
Publisher: American Geophysical Union (AGU)
Date: 12-2017
DOI: 10.1002/2017RG000581
Publisher: Wiley
Date: 28-03-2021
DOI: 10.1111/EJSS.13105
Abstract: Biochar is recommended as a soil amendment for its positive influence on soil hydrological properties, which results in improved soil fertility and crop yield. Much research in the last decade has been conducted in field and laboratory conditions on the effect of biochar on the hydraulic properties of soil. However, reported results in the literature are substantially inconsistent. Here we performed a meta‐analysis to capture the variations in change in hydraulic properties of arable soils after application of different rates of biochar. The meta‐analysis revealed that high biochar rates ( t ha −1 ) compared to low rates ( t ha −1 ) significantly improved dry bulk density in sandy and clay soils, in field and laboratory experiments. However, field capacity only improved in laboratory experiments on sandy soils. The plant available water, permanent wilting point and saturated hydraulic conductivity did not significantly increase at high rates of biochar application compared to the low rates when applied to different types of soils in both field and laboratory experiments. We discuss possible reasons for this, including hydrophobicity of the biochar with future research directions. We concluded that the current evidence does not support the notion that the application of biochar improves soils' available water capacity. Meta‐analysis clarifies the influence of biochar on soil hydraulic properties. Biochar addition at higher rates only improves the water holding capacity of sandy soils. Biochar types and pyrolysis temperatures do not influence soil hydraulic properties. The efficiency of biochar may depend on its pore size distribution and hydrophobicity.
Publisher: Elsevier BV
Date: 10-2014
Publisher: PeerJ
Date: 23-04-2013
DOI: 10.7717/PEERJ.71
Publisher: Elsevier BV
Date: 06-2016
Publisher: Elsevier BV
Date: 06-2020
Publisher: Elsevier BV
Date: 02-2018
DOI: 10.1016/J.SCITOTENV.2017.09.136
Abstract: Much research has been conducted to understand the spatial distribution of soil carbon stock and its temporal dynamics. However, an agreement has not been reached on whether increasing global temperature has a positive or negative feedback on soil carbon stocks. By analysing global maps of soil organic carbon (SOC) using a spherical wavelet analysis, it was found that the correlation between SOC and soil temperature at the regional scale was negative between 52° N and 40° S parallels and positive beyond this region. This was consistent with a few previous studies and it was assumed that the effect was most likely due to the temperature-dependent SOC formation (photosynthesis) and decomposition (microbial activities and substrate decomposability) processes. The results also suggested that the large SOC stocks distributed in the low-temperature areas might increase under global warming while the small SOC stocks found in the high-temperature areas might decrease accordingly. Although it remains unknown whether the potential increasing soil carbon stocks in the low-temperature areas can offset the loss of carbon stocks in the high-temperature areas, the location- and scale- specific correlations between SOC and temperature should be taken into account for modeling SOC dynamics and SOC sequestration management.
Publisher: Elsevier BV
Date: 04-2015
Publisher: Elsevier BV
Date: 11-2006
Publisher: Elsevier BV
Date: 03-2016
Publisher: Elsevier BV
Date: 2011
Publisher: Springer Netherlands
Date: 2011
Publisher: Elsevier BV
Date: 2011
Publisher: CSIRO Publishing
Date: 2012
DOI: 10.1071/SR12139
Abstract: The difference between the International (adopted by Australia) and the USDA/FAO particle-size classification systems is the limit between silt and sand fractions (20 μm for the International and 50 µm for the USDA/FAO). In order to work with pedotransfer functions generated under the USDA/FAO system with Australian soil survey data, a conversion should be attempted. The aim of this work is to improve prior models using larger datasets and a genetic programming technique, in the form of a symbolic regression. The 2–50 µm fraction was predicted using a USDA dataset which included both particle-size classification systems. The presented model reduced the root mean square error (%) by 14.96 and 23.62% (IGBP-DIS dataset and Australian dataset, respectively), compared with the previous model.
Publisher: Elsevier BV
Date: 08-2007
Publisher: Elsevier BV
Date: 03-2010
Publisher: MDPI AG
Date: 11-05-2022
DOI: 10.3390/SU14105815
Abstract: Recent reviews have identified major themes within regenerative agriculture—soil health, bio ersity, and socioeconomic disparities—but have so far been unable to clarify a definition based on practice and/or outcomes. In recent years, the concept has seen a rapid increase in farming, popular, and corporate interest, the scope of which now sees regenerative agriculture best viewed as a movement. To define and guide further practical and academic work in this respect, the authors have returned to the literature to explore the movement’s origins, intentions, and potential through three phases of work: early academic, current popular, and current academic. A consistent intention from early to current supporters sees the regeneration, or rebuilding, of agricultural resources, soil, water, biota, human, and energy as necessary to achieve a sustainable agriculture. This intention aligns well with international impetus to improve ecosystem function. The yet to be confirmed definition, an intention for iterative design, and emerging consumer and ecosystem service markets present several potential avenues to deliver these intentions. To assist, the authors propose the Farmscape Function framework, to monitor the impact of change in our agricultural resources over time, and a mechanism to support further data-based innovation. These tools and the movement’s intentions position regenerative agriculture as a state for rather than type of agriculture.
Publisher: Springer Science and Business Media LLC
Date: 29-04-2014
Publisher: Springer Netherlands
Date: 2008
Publisher: Elsevier BV
Date: 09-2003
Publisher: Authorea, Inc.
Date: 25-05-2023
Publisher: Elsevier BV
Date: 11-2008
Publisher: Elsevier BV
Date: 10-2022
Publisher: Elsevier BV
Date: 12-2017
Publisher: IOP Publishing
Date: 04-2021
DOI: 10.1088/1755-1315/708/1/012088
Abstract: The south-western slope of Anak Krakatau collapsed after the eruption on December 22 nd , 2018 and reshaped the volcanic island landscape. This work focused on determining the geomorphological features of Mt. Anak Krakatau before and after the eruption. A total of 71 lapilli and 17 volcanic ash s les were collected from Anak Krakatau and Panjang islands on February 23, 2019, and March 14, 2019. Sentinel-2 and Planet Scope images were utilized to monitor thermal activities and the changes of the coastlines. Google Earth Pro was capitalized to determine the rills and gullies formation. After the December 2018 eruption, the height of Anak Krakatau was reduced from 258 to 126 m and, about 76 x 10 6 m 3 of materials were eroded to the sea. The eruption caused Anak Krakatau to be covered by unconsolidated volcanic materials. About 214 of rills (dimension of 380 to 851 m and 30 to 100 cm) and 35 of the gully features (length from 150 to 841 m and width from 0.5 to 13 m) run from the highest peak to the coastline. This work can serve as a reference for predicting potentially disastrous events such as Anak Krakatau, which shows growth and destruction can be observed using remote sensing techniques.
Publisher: Springer Science and Business Media LLC
Date: 07-07-2010
Publisher: Oxford University Press (OUP)
Date: 03-2015
DOI: 10.1093/JSSAM/SMU024
Publisher: CSIRO Publishing
Date: 2016
DOI: 10.1071/CP15053
Abstract: Rainfall is a major driver for dryland wheat yields across Australia. Many authors have covered issues such as rainfall trends in Australia, and much of this information has been reviewed and updated in recent years in relation to the Millennium drought and associated concerns about climate change. However, despite a long history of work relating rainfall to grain yields, there has been no overall historical review of attempts at predictive methods and their reliability. Although many of these attempts have now been abandoned or revised, and science has moved in different directions, a review is useful to identify historical patterns and to recognise recurring themes. This might lead to new science questions and a re-appreciation of older findings. The aim of this study is therefore to review the overall literature on this topic, provide a historical timeline, and summarise the achievements and any remaining research questions. The early use of climatic data in Australia was to categorise existing and likely areas for production, with production, not surprisingly, being the emphasis. The search for a crop or climatic index was possibly initiated in an attempt to understand or simplify the complex relationships between crops and the environment. No single index has proved universally applicable, but some acceptance of early growing-season rains as an indicator seems common. The development of complex climatic models, and the availability of quality data for agricultural systems models, has allowed further quantification of the relationship between crops and climate, especially on a seasonal basis. There is little doubt that the relationship between the climatic southern oscillation phenomenon and seasonal rainfall patterns in Australia is important, but its absolute definition remains elusive. From a producer’s perspective, relationships between rainfall at specific (indicator) periods and seasonal or annual rainfall, as appropriate to specific crops, would be useful simple indicators because many farmers already maintain their own rainfall records.
Publisher: Elsevier BV
Date: 09-2007
Publisher: CSIRO Publishing
Date: 2014
DOI: 10.1071/SR13100
Abstract: An operational Digital Soil Assessment was developed to inform land suitability modelling in newly commissioned irrigation schemes in Tasmania, Australia. The Land Suitability model uses various soil parameters, along with other climate and terrain surfaces, to identify suitable areas for various agricultural enterprises for a combined 70 000-ha pilot project area in the Meander and Midlands Regions of Tasmania. An integral consideration for irrigable suitability is soil drainage. Quantitative measurement and mapping can be resource-intensive in time and associated costs, whereas more ‘traditional’ mapping approaches can be generalised, lacking the detail required for statistically validated products. The project was not sufficiently resourced to undertake replicated field-drainage measurements and relied on expert field drainage estimates at ~930 sites (260 of these for independent validation) to spatially predict soil drainage for both areas using various terrain-based and remotely sensed covariates, using three approaches: (a) decision tree spatial modelling of discrete drainage classes (b) regression-tree spatial modelling of a continuous drainage index (c) regression kriging (random-forests with residual-kriging) spatial modelling of a continuous drainage index. Method b was chosen as the best approach in terms of interpretation, and model training and validation, with a concordance coefficient of 0.86 and 0.57, respectively. A classified soil drainage map produced from the ‘index’ showed good agreement, with a linearly weighted kappa coefficient of 0.72 for training, and 0.37 for validation. The index mapping was incorporated into the overall land suitability model and proved an important consideration for the suitability of most enterprises.
Publisher: Elsevier BV
Date: 11-2014
Publisher: Wiley
Date: 05-2022
DOI: 10.1111/EJSS.13242
Abstract: We have read with interest an opinion paper recently published in the European Journal of Soil Science (Berthelin et al., 2022). This paper presents some interesting considerations, at least one of which is already well known to soil scientists working on soil organic carbon (SOC), that is, a large portion (80%–90%) of fresh carbon inputs to soil is subject to rapid mineralization. The short‐term mineralization kinetics of organic inputs is well‐known and accounted for in soil organic matter models. Thus, clearly, the long‐term predictions based on these models do not overlook short‐term mineralization. We point out that many agronomic practices can significantly contribute to SOC sequestration. If conducted responsibly whilst fully recognising the caveats, SOC sequestration can lead to a win‐win situation where agriculture can both contribute to the mitigation of climate change and adapt to it, whilst at the same time delivering other co‐benefits such as reduced soil erosion and enhanced bio ersity. Rapid mineralization of organic inputs is an important factor for soil carbon sequestration. Mineralization kinetics of organic inputs are well‐known and accounted for in soil organic matter models. Many agronomic practices can contribute significantly to SOC sequestration. SOC sequestration can lead to a win‐win situation where agriculture can both contribute to the mitigation of climate change and adapt to it.
Publisher: Elsevier BV
Date: 09-2019
Publisher: CSIRO Publishing
Date: 19-05-2021
DOI: 10.1071/SR20212
Abstract: In this study, a map of soil parent material is created to support the delineation of soil properties and classes of the Narrabri Shire, NSW. Currently, available information in this study area is geological and lithological maps at a scale of 1:250 000 to 1:1 000 000. These maps are not detailed, and the description in some areas is not accurate. Thus, this study created a new parent material map using information from the geological and lithology information, barest earth satellite imagery, gamma radiometric, topography, prior soil map and digital soil texture maps (clay and sand content). Based on interpretation and parent material observations, 18 parent material classes were delineated in the area. The 18 classes were then modelled using Linear Discriminant Analysis using Digital Elevation Model, slope, topographic wetness index, Gamma potassium (K) and thorium (Th), Ratio K to Th and soil visible and near infrared (NIR) reflectance (created using RGB and NIR bands) as covariates. This modelling process was iterated 50 times, and the most frequently predicted class was assigned to each of the 90 m × 90 m pixels throughout the study area. A map of the frequency of the predicted classes was also created to assess modelling uncertainty. The new parent material map consists of sedimentary residuals (sandstone), volcanic materials (basalt), alluvium and colluvium. The alluvium can be distinguished into six classes according to slope, soil information from satellite images and soil texture. The colluvium consists of three classes with a characteristic of high clay content (smectitic) and brown in colour (kaolinitic). Using similar approaches, such soil parent material or substrate maps could be developed for different regions in Australia. This method generated unique soil parent material classes combining stratigraphy, lithology and geomorphology.
Publisher: Wiley
Date: 20-06-2020
DOI: 10.1111/GCB.15147
Abstract: Tropical peatlands are vital ecosystems that play an important role in global carbon storage and cycles. Current estimates of greenhouse gases from these peatlands are uncertain as emissions vary with environmental conditions. This study provides the first comprehensive analysis of managed and natural tropical peatland GHG fluxes: heterotrophic (i.e. soil respiration without roots), total CO 2 respiration rates, CH 4 and N 2 O fluxes. The study documents studies that measure GHG fluxes from the soil ( n = 372) from various land uses, groundwater levels and environmental conditions. We found that total soil respiration was larger in managed peat ecosystems (median = 52.3 Mg CO 2 ha −1 year −1 ) than in natural forest (median = 35.9 Mg CO 2 ha −1 year −1 ). Groundwater level had a stronger effect on soil CO 2 emission than land use. Every 100 mm drop of groundwater level caused an increase of 5.1 and 3.7 Mg CO 2 ha −1 year −1 for plantation and cropping land use, respectively. Where groundwater is deep (≥0.5 m), heterotrophic respiration constituted 84% of the total emissions. N 2 O emissions were significantly larger at deeper groundwater levels, where every drop in 100 mm of groundwater level resulted in an exponential emission increase (exp(0.7) kg N ha −1 year −1 ). Deeper groundwater levels induced high N 2 O emissions, which constitute about 15% of total GHG emissions. CH 4 emissions were large where groundwater is shallow however, they were substantially smaller than other GHG emissions. When compared to temperate and boreal peatland soils, tropical peatlands had, on average, double the CO 2 emissions. Surprisingly, the CO 2 emission rates in tropical peatlands were in the same magnitude as tropical mineral soils. This comprehensive analysis provides a great understanding of the GHG dynamics within tropical peat soils that can be used as a guide for policymakers to create suitable programmes to manage the sustainability of peatlands effectively.
Publisher: Springer Science and Business Media LLC
Date: 28-05-2020
Publisher: Copernicus GmbH
Date: 06-02-2020
Abstract: Abstract. The application of machine learning (ML) techniques in various fields of science has increased rapidly, especially in the last 10 years. The increasing availability of soil data that can be efficiently acquired remotely and proximally, and freely available open-source algorithms, have led to an accelerated adoption of ML techniques to analyse soil data. Given the large number of publications, it is an impossible task to manually review all papers on the application of ML in soil science without narrowing down a narrative of ML application in a specific research question. This paper aims to provide a comprehensive review of the application of ML techniques in soil science aided by a ML algorithm (latent Dirichlet allocation) to find patterns in a large collection of text corpora. The objective is to gain insight into publications of ML applications in soil science and to discuss the research gaps in this topic. We found that (a) there is an increasing usage of ML methods in soil sciences, mostly concentrated in developed countries, (b) the reviewed publications can be grouped into 12 topics, namely remote sensing, soil organic carbon, water, contamination, methods (ensembles), erosion and parent material, methods (NN, neural networks, SVM, support vector machines), spectroscopy, modelling (classes), crops, physical, and modelling (continuous), and (c) advanced ML methods usually perform better than simpler approaches thanks to their capability to capture non-linear relationships. From these findings, we found research gaps, in particular, about the precautions that should be taken (parsimony) to avoid overfitting, and that the interpretability of the ML models is an important aspect to consider when applying advanced ML methods in order to improve our knowledge and understanding of soil. We foresee that a large number of studies will focus on the latter topic.
Publisher: Elsevier BV
Date: 2013
Publisher: Elsevier BV
Date: 09-2014
Publisher: Elsevier BV
Date: 2017
Publisher: Elsevier BV
Date: 10-2016
Publisher: Elsevier BV
Date: 03-2022
Publisher: Wiley
Date: 15-09-2008
Publisher: Elsevier BV
Date: 10-2019
Publisher: Elsevier BV
Date: 12-2007
Publisher: Elsevier BV
Date: 05-2008
Publisher: Wiley
Date: 04-09-2007
Publisher: Copernicus GmbH
Date: 22-03-2019
Abstract: Abstract. With the advances of new proximal soil sensing technologies, soil properties can be inferred by a variety of sensors, each having its distinct level of accuracy. This measurement error affects subsequent modelling and therefore must be integrated when calibrating a spatial prediction model. This paper introduces a deep learning model for contextual digital soil mapping (DSM) using uncertain measurements of the soil property. The deep learning model, called the convolutional neural network (CNN), has the advantage that it uses as input a local representation of environmental covariates to leverage the spatial information contained in the vicinity of a location. Spatial non-linear relationships between measured soil properties and neighbouring covariate pixel values are found by optimizing an objective function, which can be weighted with respect to a measurement error of soil observations. In addition, a single model can be trained to predict a soil property at different soil depths. This method is tested in mapping top- and subsoil organic carbon using laboratory-analysed and spectroscopically inferred measurements. Results show that the CNN significantly increased prediction accuracy as indicated by the coefficient of determination and concordance correlation coefficient, when compared to a conventional DSM technique. Deeper soil layer prediction error decreased, while preserving the interrelation between soil property and depths. The tests conducted suggest that the CNN benefits from using local contextual information up to 260 to 360 m. We conclude that the CNN is a flexible, effective and promising model to predict soil properties at multiple depths while accounting for contextual covariate information and measurement error.
Publisher: Elsevier BV
Date: 05-2013
Publisher: Elsevier BV
Date: 03-2007
Publisher: Elsevier BV
Date: 03-2019
Publisher: Informa UK Limited
Date: 31-12-2014
Publisher: Elsevier BV
Date: 2013
Publisher: Elsevier BV
Date: 11-2019
Publisher: Wiley
Date: 04-01-2022
DOI: 10.1002/CCHE.10519
Abstract: Flour millers often produce several flour types from a single wheat grist. Consequently, different specifications characterize each flour. For ex le, French standards specify six different flour types, each classified by ash content. The proportional blending of different flour streams from a single wheat grist achieves the target flour specifications. This study explores the opportunity to improve flour blending using linear programming and compares it to sequential ash curve blending. Linear programming and ash curve approaches were used to meet specifications for French flour types from a wheat grist milled to produce 10 flour streams, each stream having different flour quality attributes. The first simulation set quantity targets for Types 45, 55, and 65 flour. The balance of the flour went to the lower value Types 80, 110, and 150. The flour type targets were met using Linear Programming. By utilizing the ash curve method, Type 65 flour was under‐delivered. The second simulation aimed to maximize income using the two methods with no constraints on the amount of each flour type. The linear programming approach resulted in a 0.13% increase in revenue compared to the ash curve technique. In the first simulation, the linear programming technique reduced the lower‐value high ash flour types, generating an additional $6.16/ton of flour. In the second simulation, linear programming increased income by $0.84/ton of flour. Thus, a milling plant operating for 8,000 hr/year and processing 20 tons of wheat/hr translates to $779,000 and $107,000 per annum, respectively. This study showed that linear programming could significantly improve flour blending outcomes, resulting in increased profitability and resource utilization in the milling industry.
Publisher: Springer Science and Business Media LLC
Date: 2003
Publisher: Wiley
Date: 05-2016
Publisher: Elsevier BV
Date: 11-2015
Publisher: Elsevier
Date: 2006
Publisher: Elsevier BV
Date: 06-2020
Publisher: Wiley
Date: 2020
DOI: 10.1002/VZJ2.20060
Abstract: The “4 per 1,000” initiative calls for land management practices that increase soil organic C (SOC). Despite an imperative for accurate SOC measurement, several methodological issues may complicate the verification of C sequestration. The aim of this work is to evaluate the potential advantages of using apparent electrical conductivity (EC a )‐directed s ling to deep (0–90 cm) SOC stock assessment. We compared simple random s ling (SRS) and stratified random s ling (StSRS), with either a fixed or optimized number of s les, in fields managed under conservation agriculture and conventional tillage. The stratification in StSRS was built from EC a maps that showed two different soil conditions—the presence or absence (high‐salinity conditions) of a strong correlation between EC a and soil properties. Treatment and s ling design effects on SOC estimates were tested through a mixed‐model approach. S ling efficiency was calculated by classical and bootstrap methods. Results suggested that when EC a has a strong relationship with soil properties, StSRS was more efficient than SRS, especially when using an optimal number of s les per stratum. Stratification was based on EC a maps of the no‐till site, which allowed a smaller minimum s le size. When stratification failed due to the effect of salinity on EC a , StSRS efficiency was similar to SRS. These results suggest that EC a –directed s ling, regardless of knowing the relationships between EC a and soil properties, is a win‐win solution to advance soil characterization and SOC stock estimation in agricultural fields of the low Venetian plain. However, further research should investigate EC a –directed s ling where strong patterns not related to SOC could lead to inappropriate stratification or suboptimal s le allocation.
Publisher: PeerJ
Date: 03-10-2018
DOI: 10.7717/PEERJ.5722
Abstract: The use of visible-near infrared (vis-NIR) spectroscopy for rapid soil characterisation has gained a lot of interest in recent times. Soil spectra absorbance from the visible-infrared range can be calibrated using regression models to predict a set of soil properties. The accuracy of these regression models relies heavily on the calibration set. The optimum s le size and the overall s le representativeness of the dataset could further improve the model performance. However, there is no guideline on which s ling method should be used under different size of datasets. Here, we show different s ling algorithms performed differently under different data size and different regression models (Cubist regression tree and Partial Least Square Regression (PLSR)). We analysed the effect of three s ling algorithms: Kennard-Stone (KS), conditioned Latin Hypercube S ling (cLHS) and k-means clustering (KM) against random s ling on the prediction of up to five different soil properties (sand, clay, carbon content, cation exchange capacity and pH) on three datasets. These datasets have different coverages: a European continental dataset (LUCAS, n = 5,639), a regional dataset from Australia (Geeves, n = 379), and a local dataset from New South Wales, Australia (Hillston, n = 384). Calibration s le sizes ranging from 50 to 3,000 were derived and tested for the continental dataset and from 50 to 200 s les for the regional and local datasets. Overall, the PLSR gives a better prediction in comparison to the Cubist model for the prediction of various soil properties. It is also less prone to the choice of s ling algorithm. The KM algorithm is more representative in the larger dataset up to a certain calibration s le size. The KS algorithm appears to be more efficient (as compared to random s ling) in small datasets however, the prediction performance varied a lot between soil properties. The cLHS s ling algorithm is the most robust s ling method for multiple soil properties regardless of the s le size. Our results suggested that the optimum calibration s le size relied on how much generalization the model had to create. The use of the s ling algorithm is beneficial for larger datasets than smaller datasets where only small improvements can be made. KM is suitable for large datasets, KS is efficient in small datasets but results can be variable, while cLHS is less affected by s le size.
Publisher: PeerJ
Date: 27-10-2015
DOI: 10.7717/PEERJ.1366
Abstract: Simulations are used to generate plausible realisations of soil and climatic variables for input into an enterprise land suitability assessment (LSA). Subsequently we present a case study demonstrating a LSA (for hazelnuts) which takes into account the quantified uncertainties of the biophysical model input variables. This study is carried out in the Meander Valley Irrigation District, Tasmania, Australia. It is found that when comparing to a LSA that assumes inputs to be error free, there is a significant difference in the assessment of suitability. Using an approach that assumes inputs to be error free, 56% of the study area was predicted to be suitable for hazelnuts. Using the simulation approach it is revealed that there is considerable uncertainty about the ‘error free’ assessment, where a prediction of ‘unsuitable’ was made 66% of the time (on average) at each grid cell of the study area. The cause of this difference is that digital soil mapping of both soil pH and conductivity have a high quantified uncertainty in this study area. Despite differences between the comparative methods, taking account of the prediction uncertainties provide a realistic appraisal of enterprise suitability. It is advantageous also because suitability assessments are provided as continuous variables as opposed to discrete classifications. We would recommend for other studies that consider similar FAO (Food and Agriculture Organisation of the United Nations) land evaluation framework type suitability assessments, that parameter membership functions (as opposed to discrete threshold cutoffs) together with the simulation approach are used in concert.
Publisher: Elsevier BV
Date: 12-2020
Publisher: Elsevier BV
Date: 03-2019
Publisher: Elsevier BV
Date: 12-2019
Publisher: Elsevier BV
Date: 03-2019
Publisher: Elsevier BV
Date: 09-2017
Publisher: Elsevier BV
Date: 03-2014
Publisher: Wiley
Date: 28-10-2013
Publisher: Elsevier BV
Date: 09-2020
Publisher: Springer Science and Business Media LLC
Date: 06-08-2018
DOI: 10.1038/S41598-018-30005-8
Abstract: Soil microbial communities directly affect soil functionality through their roles in the cycling of soil nutrients and carbon storage. Microbial communities vary substantially in space and time, between soil types and under different land management. The mechanisms that control the spatial distributions of soil microbes are largely unknown as we have not been able to adequately upscale a detailed analysis of the microbiome in a few grams of soil to that of a catchment, region or continent. Here we reveal that soil microbes along a 1000 km transect have unique spatial structures that are governed mainly by soil properties. The soil microbial community assessed using Phospholipid Fatty Acids showed a strong gradient along the latitude gradient across New South Wales, Australia. We found that soil properties contributed the most to the microbial distribution, while other environmental factors (e.g., temperature, elevation) showed lesser impact. Agricultural activities reduced the variation of the microbial communities, however, its influence was local and much less than the overall influence of soil properties. The ability to predict the soil and environmental factors that control microbial distribution will allow us to predict how future soil and environmental change will affect the spatial distribution of microbes.
Publisher: Elsevier BV
Date: 2022
Publisher: Elsevier BV
Date: 07-2006
Publisher: Elsevier BV
Date: 05-2011
Publisher: MDPI AG
Date: 31-05-2019
DOI: 10.3390/SU11113072
Abstract: Volcanic eruptions affect land and humans globally. When a volcano erupts, tons of volcanic ash materials are ejected to the atmosphere and deposited on land. The hazard posed by volcanic ash is not limited to the area in proximity to the volcano, but can also affect a vast area. Ashes ejected from volcano’s affect people’s daily life and disrupts agricultural activities and damages crops. However, the positive outcome of this natural event is that it secures fertile soil for the future. This paper examines volcanic ash (tephra) from a soil security view-point, mainly its capability. This paper reviews the positive aspects of volcanic ash, which has a high capability to supply nutrients to plant, and can also sequester a large amount of carbon out of the atmosphere. We report some studies around the world, which evaluated soil organic carbon (SOC) accumulation since volcanic eruptions. The mechanisms of SOC protection in volcanic ash soil include organo-metallic complexes, chemical protection, and physical protection. Two case studies of volcanic ash from Mt. Talang and Sinabung in Sumatra, Indonesia showed the rapid accumulation of SOC through lichens and vascular plants. Volcanic ash plays an important role in the global carbon cycle and ensures soil security in volcanic regions of the world in terms of boosting its capability. However, there is also a human dimension, which does not go well with volcanic ash. Volcanic ash can severely destroy agricultural areas and farmers’ livelihoods. Connectivity and codification needs to ensure farming in the area to take into account of risk and build appropriate adaptation and resilient strategy.
Publisher: Elsevier
Date: 2021
Publisher: Elsevier BV
Date: 03-2016
Publisher: Elsevier BV
Date: 02-2020
Publisher: Elsevier BV
Date: 08-2020
Publisher: CSIRO Publishing
Date: 2006
DOI: 10.1071/SR05127
Abstract: This paper reports on a study involving the application of ultrasonic agitation to 3 soil types to assess soil aggregate disruption and subsequent dispersion. The measurement of various particle size fractions resulting after the application of ultrasonic agitation for different time periods made it possible to describe the resulting aggregate disruption using the established aggregate liberation and dispersion curve (ALDC) model. Originally this model had been used to assess only the 2–20 µm fraction liberated from Vertosols. This work has shown that the model can be applied to a variety of size fractions between 2 and 100 µm in diameter and soil types, namely Chromosols and Ferrosols. By estimating the critical energy (Ecrit) required to initiate dispersion of liberated aggregates for each fraction, it is implied that the linkage between aggregates is weaker than the linkages between the materials composing the aggregates. Further, the ratio between the rate constants in the ALDC model can be used to establish if there is a stepwise breakdown of larger aggregates, a criterion required to establish the presence of an aggregate hierarchy. Finally, by assessing the aggregate distribution on a continuous scale, it is possible to recognise unique pathways of aggregate liberation and dispersion for each soil type rather than assuming that aggregates breakdown into predefined discrete size fractions.
Publisher: Wiley
Date: 07-2007
Publisher: Elsevier BV
Date: 2022
Publisher: Informa UK Limited
Date: 02-2006
Publisher: Elsevier BV
Date: 03-2017
Publisher: Springer Science and Business Media LLC
Date: 06-01-2010
Publisher: Elsevier BV
Date: 06-2021
Publisher: Elsevier BV
Date: 02-2012
Publisher: Springer Science and Business Media LLC
Date: 26-06-2009
Publisher: Elsevier BV
Date: 11-2011
Publisher: Elsevier BV
Date: 2018
Publisher: Elsevier BV
Date: 03-2010
Publisher: Elsevier BV
Date: 06-2018
Publisher: Elsevier BV
Date: 07-2021
Publisher: Elsevier BV
Date: 05-2016
Publisher: Elsevier BV
Date: 04-2023
Publisher: Elsevier BV
Date: 11-2016
Publisher: Elsevier BV
Date: 06-2020
Publisher: Wiley
Date: 09-2002
Publisher: Elsevier BV
Date: 07-2006
Publisher: Elsevier BV
Date: 11-2003
Publisher: Wiley
Date: 09-2022
DOI: 10.1111/EJSS.13285
Abstract: Since the early 2000s, digital soil maps have been successfully used for various applications, including precision agriculture, environmental assessments and land use management. Globally, however, there are large disparities in the availability of soil data on which digital soil mapping (DSM) models can be fitted. Several studies attempted to transfer a DSM model fitted from an area with a well‐developed soil database to map the soil in areas with low s ling density. This usually is a challenging task because two areas have hardly ever the same soil‐forming factors in two different regions of the world. In this study, we aim to determine whether finding homosoils (i.e., locations sharing similar soil‐forming factors) can help transferring soil information by means of a DSM model extrapolation. We hypothesize that within areas in the world considered as homosoils, one can leverage on areas with high s ling density and fit a DSM model, which can then be extrapolated geographically to an area with little or no data. We collected publicly available soil data for clay, silt, sand, organic carbon (OC), pH and total nitrogen (N) within our study area in Mali, West Africa and its homosoils. We fitted a regression tree model between the soil properties and environmental covariates of the homosoils, and applied this model to our study area in Mali. Several calibration and validation strategies were explored. We also compared our approach with existing maps made at a global and a continental scale. We concluded that geographic model extrapolation within homosoils was possible, but that model accuracy dramatically improved when local data were included in the calibration dataset. The maps produced from models fitted with data from homosoils were more accurate than existing products for this study area, for three (silt, sand, pH) out of six soil properties. This study would be relevant to areas with very little or no soil data to carry critical soils and environmental risk assessments at a regional level. Soil mapping models were fitted with soil data within the homosoils of Mali. The fitted models were applied to our study area. Model accuracy dramatically improved when including local data. Homosoil maps were more accurate for 3 out of 6 soil properties compared to global and continental maps. New opportunity to map the regional soil pattern of areas with limited soil data coverage.
Publisher: PeerJ
Date: 07-10-2020
DOI: 10.7717/PEERJ.10106
Abstract: Surface air temperature ( T a ) required for real-time environmental modelling applications should be spatially quantified to capture the nuances of local-scale climates. This study created near real-time air temperature maps at a high spatial resolution across Australia. This mapping is achieved using the thin plate spline interpolation in concert with a digital elevation model and ‘live’ recordings garnered from 534 telemetered Australian Bureau of Meteorology automatic weather station (AWS) sites. The interpolation was assessed using cross-validation analysis in a 1-year period using 30-min interval observation. This was then applied to a fully automated mapping system—based in the R programming language—to produce near real-time maps at sub-hourly intervals. The cross-validation analysis revealed broad similarities across the seasons with mean-absolute error ranging from 1.2 °C (autumn and summer) to 1.3 °C (winter and spring), and corresponding root-mean-square error in the range 1.6 °C to 1.7 °C. The R 2 and concordance correlation coefficient ( P c ) values were also above 0.8 in each season indicating predictions were strongly correlated to the validation data. On an hourly basis, errors tended to be highest during the late afternoons in spring and summer from 3 pm to 6 pm, particularly for the coastal areas of Western Australia. The mapping system was trialled over a 21-day period from 1 June 2020 to 21 June 2020 with majority of maps completed within 28-min of AWS site observations being recorded. All outputs were displayed in a web mapping application to exemplify a real-time application of the outputs. This study found that the methods employed would be highly suited for similar applications requiring real-time processing and delivery of climate data at high spatiotemporal resolutions across a considerably large land mass.
Publisher: American Association for the Advancement of Science (AAAS)
Date: 07-08-2009
Abstract: Increased demand and advanced techniques could lead to more refined mapping and management of soils.
Publisher: Elsevier BV
Date: 03-2008
Publisher: CSIRO Publishing
Date: 2006
DOI: 10.1071/SR05136
Abstract: Estimation and mapping carbon storage in the soil is currently an important topic thus, the knowledge of the distribution of carbon content with depth is essential. This paper examines the use of a negative exponential profile depth function to describe the soil carbon data at different depths, and its integral to represent the carbon storage. A novel method is then proposed for mapping the soil carbon storage in the Lower Namoi Valley, NSW. This involves deriving pedotransfer functions to predict soil organic carbon and bulk density, fitting the exponential depth function to the carbon profile data, deriving a neural network model to predict parameters of the exponential function from environmental data, and mapping the organic carbon storage. The exponential depth function is shown to fit the soil carbon data adequately, and the parameters also reflect the influence of soil order. The parameters of the exponential depth function were predicted from land use, radiometric K, and terrain attributes. Using the estimated parameters we map the carbon storage of the area from surface to a depth of 1 m. The organic carbon storage map shows the high influence of land use on the predicted storage. Values of 15–22 kg/m2 were predicted for the forested area and 2–6 kg/m2 in the cultivated area in the plains.
Publisher: Ovid Technologies (Wolters Kluwer Health)
Date: 08-2013
Publisher: Elsevier BV
Date: 11-2014
Publisher: Elsevier BV
Date: 09-2021
Publisher: Elsevier
Date: 2020
Publisher: Springer Science and Business Media LLC
Date: 08-05-2010
Publisher: Springer Science and Business Media LLC
Date: 12-12-2022
Publisher: Elsevier BV
Date: 04-2017
Publisher: Elsevier
Date: 2006
Publisher: Elsevier BV
Date: 12-2006
Publisher: Elsevier BV
Date: 10-2005
Publisher: Elsevier BV
Date: 02-2014
Publisher: Elsevier BV
Date: 02-2014
Publisher: Elsevier BV
Date: 11-2022
Publisher: Elsevier BV
Date: 03-2018
Publisher: Springer Science and Business Media LLC
Date: 04-02-2017
DOI: 10.1007/S11356-017-8511-X
Abstract: The distribution of heavy metals in agricultural soils is affected by various anthropogenic activities and environmental factors occurring at different spatial scales. This paper introduced the two-dimensional empirical mode decomposition (2D-EMD) to separate the spatial variability in soil heavy metals into different scales. Geostatistics and multivariate analysis were also utilized to quantify their spatial structure and identify their potential influencing factors. The study was conducted in an arable land in southeastern China where 260 surface soil s les were collected and measured for total contents of cadmium (Cd
Publisher: Elsevier BV
Date: 09-2019
Publisher: Elsevier BV
Date: 11-2017
Publisher: PeerJ
Date: 21-07-2022
DOI: 10.7717/PEERJ.13740
Abstract: Improving the amount of organic carbon in soils is an attractive alternative to partially mitigate climate change. However, the amount of carbon that can be potentially added to the soil is still being debated, and there is a lack of information on additional storage potential on global cropland. Soil organic carbon (SOC) sequestration potential is region-specific and conditioned by climate and management but most global estimates use fixed accumulation rates or time frames. In this study, we model SOC storage potential as a function of climate, land cover and soil. We used 83,416 SOC observations from global databases and developed a quantile regression neural network to quantify the SOC variation within soils with similar environmental characteristics. This allows us to identify similar areas that present higher SOC with the difference representing an additional storage potential. We estimated that the topsoils (0–30 cm) of global croplands (1,410 million hectares) hold 83 Pg C. The additional SOC storage potential in the topsoil of global croplands ranges from 29 to 65 Pg C. These values only equate to three to seven years of global emissions, potentially offsetting 35% of agriculture’s 85 Pg historical carbon debt estimate due to conversion from natural ecosystems. As SOC store is temperature-dependent, this potential is likely to reduce by 14% by 2040 due to climate change in a “business as usual” scenario. The results of this article can provide a guide to areas of focus for SOC sequestration, and highlight the environmental cost of agriculture.
Publisher: Elsevier BV
Date: 09-2007
Publisher: Elsevier
Date: 2004
Publisher: Wiley
Date: 2018
DOI: 10.1111/EJSS.12526
Publisher: Elsevier BV
Date: 12-2007
Publisher: Elsevier BV
Date: 06-2017
Publisher: Springer Science and Business Media LLC
Date: 10-10-2023
Publisher: Wiley
Date: 11-2010
Publisher: Elsevier BV
Date: 04-2016
Publisher: Elsevier
Date: 2011
Publisher: Public Library of Science (PLoS)
Date: 19-08-2014
Publisher: Elsevier BV
Date: 05-2015
Publisher: Elsevier BV
Date: 10-2015
Publisher: PeerJ
Date: 12-02-2013
DOI: 10.7717/PEERJ.6
Publisher: CSIRO Publishing
Date: 2000
DOI: 10.1071/SR99110
Abstract: Pedotransfer functions (PTFs) for predicting saturated hydraulic conductivity (Ks) were evaluated using published Australian soil data sets. Eight published PTFs were evaluated. Generally, published PTFs provide a satisfactory estimation of Ks depending on the spatial scale and accuracy of prediction. Several PTFs were developed in this study, including the power function of effective porosity, multiple linear regression, fractal model, and artificial neural networks. Different methods for estimating the fractal dimension of particle-size distributions showed no significant differences in predicting Ks . The simplest model for estimating fractal dimension from the log–log plot of particle-size distribution is therefore recommended. The data set was also stratified into 3 broad classes of texture: sandy, loamy, and clayey. Stratification of PTFs based on textural class showed small improvements in estimation. The published PTF of Dane and Puckett (1994) Proc. Int. Workshop (Univ. of California: Riverside, CA) gives the best prediction for sandy soil the PTF of Cosby et al. (1984) Water Resources Research 20, 682–90 gives the best production for loamy soil and the PTF of Schaap et al. (1998) Soil Science Society of America Journal 62, 847–55 gives the best prediction for clayey soil. The data set used comprised different field and laboratory measurements over large areas, and limited predictive variables were available. The PTFs developed here may predict adequately in large areas (residuals = 10–20 mm/h), but for site-specific applications, local calibration is needed.
Publisher: Elsevier BV
Date: 10-2009
Publisher: Elsevier BV
Date: 04-2017
Publisher: MDPI AG
Date: 06-02-2019
Abstract: Crop production needs to double to feed the world’s growing population. Indonesia, as the fourth most populated country in the world, needs to meet its food security challenge with a shrinking arable land area. Indonesia has over 34 million ha of sw land. The scarcity of arable land in Indonesia means wetlands are likely to be converted to agricultural use. The challenge is to both profitably and sustainably do so. This paper presents a framework for developing wetlands for food production, which includes (1) the characterization of land and problem of development (2) analysis of historical development and lessons learned (3) technology development and (4) optimization of development. We analyze each of the components and its relation to regional economic growth and lessons learned. For successful future wetland development, three factors must be considered: Land-soil-water characterization, landscape and land use design, and community development. This framework can be adopted by other tropical areas for the development of wetlands.
Publisher: Agrivita, Journal of Agricultural Science (AJAS)
Date: 06-2020
Publisher: CSIRO Publishing
Date: 2001
DOI: 10.1071/SR99114
Abstract: A novel form of ordinary kriging, involving the local estimation and modelling of the variogram at each prediction site (OKLV), is tested at a regional scale on a large data set, in order to adapt to non-uniform spatial structures and improve the assessment of the salinity hazard in the lower Chelif Valley, Algeria. The spatial variability study was carried out on a 38000 ha area using 5141 topsoil electrical conductivity (EC) measurements systematically s led on a 250 m by 250 m grid. Variography analysis confirmed the existence of large trends in the EC variability with differing spatial structures between sub-areas. OKLV performed better than ordinary kriging with a whole-area variogram (OKWV) in predicting the proportion of high saline soils in large blocks, but the predictions appeared mostly similar. In contrast, the estimation variance maps revealing the uncertainties of the spatial predictions were markedly different between the 2 methods. OKLV integrates the local spatial structure in the uncertainty assessment, whereas kriging with a whole-area variogram only considers the s ling intensity. Comparison with prediction errors on a validation set confirmed the consistency of the OKLV prediction variance. This appears to be a major improvement for decision-making procedures such as delineating areas where remediation should take place.
Publisher: Elsevier BV
Date: 09-2016
Publisher: American Geophysical Union (AGU)
Date: 02-04-2013
DOI: 10.1029/2011JF002296
Publisher: MDPI AG
Date: 05-02-2022
DOI: 10.3390/RS14030740
Abstract: Although many Soil Spectral Libraries (SSLs) have been created globally, these libraries still have not been operationalized for end-users. To address this limitation, this study created an online Brazilian Soil Spectral Service (BraSpecS). The system was based on the Brazilian Soil Spectral Library (BSSL) with s les collected in the Visible–Near–Short-wave infrared (vis–NIR–SWIR) and Mid-infrared (MIR) ranges. The interactive platform allows users to find spectra, act as custodians of the data, and estimate several soil properties and classification. The system was tested by 500 Brazilian and 65 international users. Users accessed the platform (besbbr.com.br), uploaded their spectra, and received soil organic carbon (SOC) and clay content prediction results via email. The BraSpecS prediction provided good results for Brazilian data, but performed variably for other countries. Prediction for countries outside of Brazil using local spectra (External Country Soil Spectral Libraries, ExCSSL) mostly showed greater performance than BraSpecS. Clay R2 ranged from 0.5 (BraSpecS) to 0.8 (ExCSSL) in vis–NIR–SWIR, but BraSpecS MIR models were more accurate in most situations. The development of external models based on the fusion of local s les with BSSL formed the Global Soil Spectral Library (GSSL). The GSSL models improved soil properties prediction for different countries. Nevertheless, the proposed system needs to be continually updated with new spectra so they can be applied broadly. Accordingly, the online system is dynamic, users can contribute their data and the models will adapt to local information. Our community-driven web platform allows users to predict soil attributes without learning soil spectral modeling, which will invite end-users to utilize this powerful technique.
Publisher: Springer Science and Business Media LLC
Date: 29-12-2020
Publisher: CSIRO Publishing
Date: 2010
DOI: 10.1071/SR09111
Abstract: The reliable assessment of soil carbon stock is of key importance for soil conservation and mitigation strategies related to reducing atmospheric carbon. Measuring and monitoring soil carbon is complex because carbon pools cycle and rates of carbon sequestration vary across the landscape due to climate, soil type, and management practices. A new methodology has been developed and applied to make an assessment of the distribution of total, organic, and inorganic carbon at a grains research and grazing property in northern New South Wales at a high spatial resolution. In this study, baseline soil carbon maps were created using fine resolution, geo-referenced, proximal sensor data. Coupled with a digital elevation model and secondary terrain attributes, all of the data layers were combined by k-means clustering to develop a stratified random soil s ling scheme for the survey area. Soil s les taken at 0.15-m increments to a depth of 1 m were scanned with a mid-infrared spectrometer, which was calibrated using a proportion of the s les that were analysed in a laboratory for total carbon and inorganic carbon content. This combination of new methodologies and technologies has the potential to provide large volumes of reliable, fine resolution and timely data required to make baseline assessments, mapping, monitoring, and verification possible. This method has the potential to make soil carbon management and trading at the farm-scale possible by quantifying the carbon stock to a depth of 1 m and at a high spatial resolution.
Publisher: Elsevier BV
Date: 2018
Publisher: Elsevier BV
Date: 02-2012
Publisher: Elsevier BV
Date: 02-2021
Publisher: Elsevier BV
Date: 06-2011
Publisher: Elsevier BV
Date: 06-2011
Publisher: Elsevier BV
Date: 2022
Publisher: Wiley
Date: 16-08-2011
Publisher: Elsevier BV
Date: 07-2011
Publisher: Elsevier
Date: 2013
Publisher: Elsevier BV
Date: 03-2018
Publisher: Springer Science and Business Media LLC
Date: 18-12-2012
Publisher: CSIRO Publishing
Date: 2021
DOI: 10.1071/SR19395
Abstract: Bulk density and soil stiffness moduli are vital physical parameters related to soil compaction, porosity, moisture storage capacity, soil penetration resistance and structural integrity. Conventional methods for measuring soil density and stiffness moduli are destructive, time-consuming, complex, expensive and often require skilled operators to conduct the tests. A new soil density and stiffness moduli measurement technique that can evaluate soil density and stiffness moduli more rapidly, efficiently and precisely, at a low cost is introduced here. This study evaluated the use of shear wave velocity measurements using the piezoelectric extender and bender elements as a viable alternative to measure soil density and stiffness moduli of soil. To test this idea, soda-lime glass beads of & .002, 0.04–0.07 and 1.00–1.30 mm in diameter were used to develop the empirical relationship between the shear wave velocity and the bulk density of soil in laboratory conditions. These empirical equations were then tested on sands and clayey soils for validation. Accuracy in terms of coefficient of determination (R2) and root mean squared error (RMSE) from the current and existing studies ranged within 0.91–0.93 and 0.073–0.177 g cm–3 respectively. Both shear and Young moduli were compared with the shear wave velocity of soil, with R2 and RMSE of 0.96–0.97 and 0.48–3.5 MPa respectively. The major advantage of this technique is that input and output signal data can be stored in a computer that can be used to calculate soil density and stiffness moduli automatically. This technique could play a vital role in improving crop yield and soil management practices.
Publisher: CSIRO Publishing
Date: 2015
DOI: 10.1071/SR14268
Abstract: Until recently, Tasmanian environmental modelling and assessments requiring important soil inputs relied on conventionally derived soil polygons that were mapped up to 75 years ago. In the ‘Wealth from Water’ project, digital soil mapping (DSM) was used in a pilot project to map the suitability of 20 different agricultural enterprises over 70 000 ha. Following on from this, the Tasmanian Department of Primary Industries Parks Water and Environment has applied DSM to existing soil datasets to develop enterprise suitability predictions across the whole state in response to further expansion of irrigation schemes. The soil surfaces generated have conformed and contributed to the Terrestrial Ecosystem Research Network Soil and Landscape Grid of Australia, a superset of GlobalSoilMap.net specifications. The surfaces were generated at 80-m resolution for six standard depths and 13 soil properties (e.g. pH, EC, organic carbon, sand and silt percentages and coarse fragments), in addition to several Tasmanian enterprise-suitability soil-attribute parameters. The modelling used soil site data with available explanatory state-wide spatial variables, including the Shuttle Radar Topography Mission digital elevation model and derivatives, gamma-radiometrics, surface geology, and multi-spectral satellite imagery. The DSM has delivered realistic mapping for most attributes, with acceptable validation diagnostics and relatively low uncertainty ranges in data-rich areas, but performed marginally in terms of uncertainty ranges in areas such as the World Heritage-listed Southwest of the state, with a low existing soil site density. Version 1.0 soil-attribute maps form the foundations of a dynamic and evolving new infrastructure that will be improved and re-run with the future collection of new soil data. The Tasmanian mapping has provided a localised integration with the National Soil and Landscape Grid of Australia, and it will guide future investment in soil information capture by quantitatively targeting areas with both high uncertainties and important ecological or agricultural value.
Publisher: Elsevier BV
Date: 07-2010
Publisher: Brill
Date: 25-06-2022
DOI: 10.1163/22134379-BJA10037
Abstract: This article traces one narrative of anti-colonial violence on the Sumatra plantation through various Sinophone iterations and establishes the historical events on which it was based. The European anxiety about the defiance of the condemned Chinese men shows how this particular event turned into oral legend, religious observance, touring socialist theatre, leftist fiction, and a PRC Third World internationalist travelogue. In one moment of bravura, Chinese plantation workers rejected their status as colonial subjects. That gesture made them an emblem of the proletarian bona fides of the ethnic Chinese in Indonesia, and of the traumatic origins of Medan and other North Sumatra Chinese communities in plantation labour. By connecting the foreboding in the colonial archive with the eulogy in the Sinophone literary record, we can triangulate a fuller vision of resistance on the Deli plantations than is available from either one.
Publisher: Elsevier BV
Date: 08-2018
DOI: 10.1016/J.SCITOTENV.2018.02.302
Abstract: Spatial modelling of environmental data commonly only considers spatial variability as the single source of uncertainty. In reality however, the measurement errors should also be accounted for. In recent years, infrared spectroscopy has been shown to offer low cost, yet invaluable information needed for digital soil mapping at meaningful spatial scales for land management. However, spectrally inferred soil carbon data are known to be less accurate compared to laboratory analysed measurements. This study establishes a methodology to filter out the measurement error variability by incorporating the measurement error variance in the spatial covariance structure of the model. The study was carried out in the Lower Hunter Valley, New South Wales, Australia where a combination of laboratory measured, and vis-NIR and MIR inferred topsoil and subsoil soil carbon data are available. We investigated the applicability of residual maximum likelihood (REML) and Markov Chain Monte Carlo (MCMC) simulation methods to generate parameters of the Matérn covariance function directly from the data in the presence of measurement error. The results revealed that the measurement error can be effectively filtered-out through the proposed technique. When the measurement error was filtered from the data, the prediction variance almost halved, which ultimately yielded a greater certainty in spatial predictions of soil carbon. Further, the MCMC technique was successfully used to define the posterior distribution of measurement error. This is an important outcome, as the MCMC technique can be used to estimate the measurement error if it is not explicitly quantified. Although this study dealt with soil carbon data, this method is amenable for filtering the measurement error of any kind of continuous spatial environmental data.
Publisher: Pleiades Publishing Ltd
Date: 28-09-2021
Publisher: Elsevier BV
Date: 12-2021
Publisher: Elsevier BV
Date: 09-2022
Publisher: Elsevier BV
Date: 05-2002
Publisher: Elsevier BV
Date: 2018
Publisher: Elsevier BV
Date: 2017
Publisher: Elsevier BV
Date: 02-2016
Publisher: Elsevier BV
Date: 02-2019
Publisher: CSIRO Publishing
Date: 2009
DOI: 10.1071/SR09005
Abstract: This paper demonstrates the application of near infrared diffuse reflectance spectroscopy (NIR-DRS) measurements as part of digital soil mapping. We also investigate whether calibration functions developed from a spectral library can be used for rapid characterisation of soil properties in the field. Soil s les were collected along 24 toposequences in the Pokolbin irrigation district, ~7 km2 of predominantly agricultural land in the Hunter Valley, NSW, Australia. Soil s les at 2 depths: 0–0.10 and 0.40–0.50 m were collected. The soil s les were scanned using NIR under 3 different conditions: field condition, dried unground, and dried ground. A separate spectral library containing soil laboratory measurements was used to develop functions to predict 3 main soil properties from NIR spectra (total C content, clay content, and sum of exchangeable cations). The absorbance spectra were found to be different for the 3 soil conditions. The field spectra appear to have higher absorbance, followed by dried unground s les and then dried ground s les. Although most spectral signatures or peaks were similar for the 3 soil conditions, field s les appear to have higher absorbance, particularly at 1400 nm and 1900 nm. The convex hull of the first 2 principal components of the soil spectra is an easy tool to evaluate the similarity of spectra from a calibration set to an observation. For field prediction, s les need to be calibrated using field s les. Finally, this study shows that NIR-DRS measurement is a useful part of digital soil mapping.
Publisher: Elsevier BV
Date: 06-2022
Publisher: American Geophysical Union (AGU)
Date: 04-2019
DOI: 10.1029/2018WR022797
Publisher: Copernicus GmbH
Date: 26-02-2019
Abstract: Abstract. Digital soil mapping (DSM) has been widely used as a cost-effective method for generating soil maps. However, current DSM data representation rarely incorporates contextual information of the landscape. DSM models are usually calibrated using point observations intersected with spatially corresponding point covariates. Here, we demonstrate the use of the convolutional neural network (CNN) model that incorporates contextual information surrounding an observation to significantly improve the prediction accuracy over conventional DSM models. We describe a CNN model that takes inputs as images of covariates and explores spatial contextual information by finding non-linear local spatial relationships of neighbouring pixels. Unique features of the proposed model include input represented as a 3-D stack of images, data augmentation to reduce overfitting, and the simultaneous prediction of multiple outputs. Using a soil mapping ex le in Chile, the CNN model was trained to simultaneously predict soil organic carbon at multiples depths across the country. The results showed that, in this study, the CNN model reduced the error by 30 % compared with conventional techniques that only used point information of covariates. In the ex le of country-wide mapping at 100 m resolution, the neighbourhood size from 3 to 9 pixels is more effective than at a point location and larger neighbourhood sizes. In addition, the CNN model produces less prediction uncertainty and it is able to predict soil carbon at deeper soil layers more accurately. Because the CNN model takes the covariate represented as images, it offers a simple and effective framework for future DSM models.
Publisher: Informa UK Limited
Date: 10-01-2022
Publisher: MDPI AG
Date: 12-07-2019
DOI: 10.3390/RS11141666
Abstract: More than 50% of the world’s population consumes rice. Accurate and up-to-date information on rice field extent is important to help manage food and water security. Currently, field surveys or MODIS satellite data are used to estimate rice growing areas. This study presents a cost-effective methodology for near-real-time mapping and monitoring of rice growth extent and cropping patterns over a large area. This novel method produces high-resolution monthly maps (10 m resolution) of rice growing areas, as well as rice growth stages. The method integrates temporal Sentinel-1 data and rice phenological parameters with the Google Earth Engine (GEE) cloud-based platform. It uses monthly median time series of Sentinel-1 at VH polarization from September 2016 to October 2018. The two study areas are the northern region of West Java, Indonesia (0.75 million ha), and the Kedah and Perlis states in Malaysia (over 1 million ha). K-means clustering, hierarchical cluster analysis (HCA), and a visual interpretation of VH polarization time series profiles are used to generate rice extent, cropping patterns, and spatiotemporal distribution of growth stages. To automate the process, four supervised classification methods (support vector machine (SVM), artificial neural networks (ANN), random forests, and C5.0 classification models) were independently trialled to identify cluster labels. The results from each classification method were compared. The method can also forecast rice extent for up to two months. The VH polarization data can identify four growth stages of rice—T& P: tillage and planting (30 days) V: vegetative-1 and 2 (60 days) R: reproductive (30 days) M: maturity (30 days). Compared to field survey data, this method measures overall rice extent with an accuracy of 96.5% and a kappa coefficient of 0.92. SVM and ANN show better performance than random forest and C5.0 models. This simple and robust method could be rolled out across Southeast Asia, and could be used as an alternative to time-consuming, expensive field surveys.
Publisher: Elsevier BV
Date: 2014
Publisher: Elsevier BV
Date: 12-2020
Publisher: Elsevier BV
Date: 12-2018
Publisher: Elsevier BV
Date: 03-2009
Publisher: Wiley
Date: 03-2019
DOI: 10.1111/EJSS.12790
Publisher: IOP Publishing
Date: 05-2020
DOI: 10.1088/1755-1315/504/1/012020
Abstract: Fires on tropical peatlands in Indonesia are unexpected events that happened sporadically, especially during dry seasons. It is not easy to judge why a fire presents in a certain spot, and how it spreads. There are indications, and originally, a fire is ignited intentionally to clear bushes and wood remnants on the land surface for land clearing. Its most negative impacts are on human health due to dispersing smokes locally, and occasionally, reached neighboring countries. It is not unusual, the fire spread beyond the initial spot, and burned a vast area of commercial plantations, and protected forests in the vicinity. Fires caused significant economic losses. This investigation aimed at making clearer how peat properties changed before and sometime after the fires burned the areas. We observed how fires in peatlands occurred, and to what extent the physical properties of peats altered due to fires. Observations were carried out in 4 locations within Riau province, South Sumatra province, and South Kalimantan province that have vast areas of peatlands in the country, and peatland fires there are more frequent. The observations focused on peatland morphological appearances, and physical properties of non-burned and after fires in protected forests, plantations, and smallholder lands. The physical and hydraulic properties that were analyzed, among others, are bulk density, particle density, porosity, available water capacity, and water permeability. Positive changes were experienced successively by the air-entry head (21.3%) followed by permeability (19.6%), particle density (15.9%), bulk density (10.7%), available water (7.4%), fast-drain pore (3.9%) and n-parameter (0.1%). While, negative changes were experienced successively by slow-drain pore (-72.2%), residual water content (-22.5%), porosity (-7.3%), saturated water content (-6.3%) and m-parameter (-2.6%). The increase of air-entry head and the available water indicated more capable of retaining more water. Meanwhile, the increase of permeability and fast pore drain mean the burnt peats would be easier to drain under the same suction gradient. However, based on the F-test with a probability value of 5%, those all changes were not so significant or considerably small compared to the associated critical values.
Publisher: Elsevier BV
Date: 05-2012
Publisher: Springer Science and Business Media LLC
Date: 2003
Publisher: Elsevier BV
Date: 10-2015
Publisher: Elsevier BV
Date: 12-1999
Publisher: Elsevier BV
Date: 05-2020
Publisher: Elsevier BV
Date: 09-2020
Publisher: Copernicus GmbH
Date: 17-11-2020
Abstract: Abstract. The number of s les used in the calibration data set affects the quality of the generated predictive models using visible, near and shortwave infrared (VIS–NIR–SWIR) spectroscopy for soil attributes. Recently, the convolutional neural network (CNN) has been regarded as a highly accurate model for predicting soil properties on a large database. However, it has not yet been ascertained how large the s le size should be for CNN model to be effective. This paper investigates the effect of the training s le size on the accuracy of deep learning and machine learning models. It aims at providing an estimate of how many calibration s les are needed to improve the model performance of soil properties predictions with CNN as compared to conventional machine learning models. In addition, this paper also looks at a way to interpret the CNN models, which are commonly labelled as a black box. It is hypothesised that the performance of machine learning models will increase with an increasing number of training s les, but it will plateau when it reaches a certain number, while the performance of CNN will keep improving. The performances of two machine learning models (partial least squares regression – PLSR Cubist) are compared against the CNN model. A VIS–NIR–SWIR spectra library from Brazil, containing 4251 unique sites with averages of two to three s les per depth (a total of 12 044 s les), was ided into calibration (3188 sites) and validation (1063 sites) sets. A subset of the calibration data set was then created to represent a smaller calibration data set ranging from 125, 300, 500, 1000, 1500, 2000, 2500 and 2700 unique sites, which is equivalent to a s le size of approximately 350, 840, 1400, 2800, 4200, 5600, 7000 and 7650. All three models (PLSR, Cubist and CNN) were generated for each s le size of the unique sites for the prediction of five different soil properties, i.e. cation exchange capacity, organic carbon, sand, silt and clay content. These calibration subset s ling processes and modelling were repeated 10 times to provide a better representation of the model performances. Learning curves showed that the accuracy increased with an increasing number of training s les. At a lower number of s les ( 1000), PLSR and Cubist performed better than CNN. The performance of CNN outweighed the PLSR and Cubist model at a s le size of 1500 and 1800, respectively. It can be recommended that deep learning is most efficient for spectra modelling for s le sizes above 2000. The accuracy of the PLSR and Cubist model seems to reach a plateau above s le sizes of 4200 and 5000, respectively, while the accuracy of CNN has not plateaued. A sensitivity analysis of the CNN model demonstrated its ability to determine important wavelengths region that affected the predictions of various soil attributes.
Publisher: Elsevier BV
Date: 02-2022
DOI: 10.1016/J.SCITOTENV.2021.152086
Abstract: Anthropogenic activities, in addition to climate change caused the drying of Urmia Lake in Iran, since 2005. Dust storms blown from the dried lakebed have created serious environmental hazards in adjacent areas. These crises would jeopardise achieving United Nations Sustainable Development Goals (UN SDGs) and emphasise the need for evaluating the spatial distribution of soil enrichment of potentially toxic elements (PTEs) (As, Cr, Cu, Ni, Pb and Zn). Conventional assessment would require a costly s ling method to map potentially polluted areas. Digital soil mapping (DSM) has proved to be a cost-efficient method for soil mapping, however its application in mapping enrichment of PTEs in soil is still lacking. This study aims to map and project the potential pollution of PTEs in the Urmia Lake area using digital mapping techniques and Landsat-8 OLI satellite images. A total of 129 surficial soil s les were collected as ground control. Enrichment factors (EFs) of PTEs and the Modified Pollution Index (MPI) were spatially predicted using two machine learning models. Covariates were derived from a suite of Landsat-8 spectral indices. The bootstrapping method was used to analyse the uncertainties. The results showed that Random Forests performed well in estimating EFs of several PTEs. Spectral indices using NIR and SWIR bands were key to predict these PTEs and MPI. The digital maps demonstrated that the study area was enriched with As, Cu and Pb at moderate to significant levels. Regions under the lower ecological level (elevation <-1274 m) had significantly larger enrichment than those of higher elevation. Based on MPI, 43% of the area was categorised as moderately polluted, and 31% of the area was moderately-heavily polluted. Possible sources of PTEs were discharges from farmlands, landfills, and industries. Our results revealed that the Urmia Lake desiccating has caused severe environmental challenges and needs immediate restoration.
Publisher: Wiley
Date: 12-2018
DOI: 10.1111/SUM.12463
Publisher: MDPI AG
Date: 14-01-2023
DOI: 10.3390/LAND12010255
Abstract: Soil organic carbon (SOC) storage and redistribution across the landscape (through erosion and deposition) are linked to soil physicochemical properties and can affect soil quality. However, the spatial and temporal variability of soil erosion and SOC remains uncertain. Whether soil redistribution leads to SOC gains or losses continues to be hotly debated. These considerations cannot be modelled using conventional soil carbon models and digital soil mapping. This paper presents a coupled-model combining RothPC-1 which considers soil carbon (C) down to 1 m and a soil redistribution model. The soil redistribution component is based on a cellular automata technique using the multi-direction flow (FD8) algorithm. With the optimized input values based on land use, we simulated SOC changes upon soil profiles to 1 m across the Lower Hunter Valley area (11,300 ha) in New South Wales, Australia from the 1970s to 2016. Results were compared to field observations and showed that erosion was predicted mostly in upslope areas and deposition in low-lying areas. We further simulated SOC trends from 2017 until ~2045 in the area under three climate scenarios and five land use projections. The variation in the magnitude and direction of SOC change with different projections shows that the main factors influencing SOC changes considering soil redistribution are climate change which controlled the trend of SOC stocks, followed by land use change. Neglecting soil erosion in carbon models could lead to an overestimation of SOC stocks. This paper provides a framework for incorporating soil redistribution into the SOC dynamics modelling and also postulates the thinking that soil erosion is not just a removal process by surface runoff.
Publisher: Springer Science and Business Media LLC
Date: 27-10-2020
DOI: 10.1038/S41467-020-18887-7
Abstract: Sustainable soil carbon sequestration practices need to be rapidly scaled up and implemented to contribute to climate change mitigation. We highlight that the major potential for carbon sequestration is in cropland soils, especially those with large yield gaps and/or large historic soil organic carbon losses. The implementation of soil carbon sequestration measures requires a erse set of options, each adapted to local soil conditions and management opportunities, and accounting for site-specific trade-offs. We propose the establishment of a soil information system containing localised information on soil group, degradation status, crop yield gap, and the associated carbon-sequestration potentials, as well as the provision of incentives and policies to translate management options into region- and soil-specific practices.
Publisher: PeerJ
Date: 22-10-2013
DOI: 10.7717/PEERJ.183
Publisher: Wiley
Date: 25-03-2013
Publisher: Elsevier
Date: 2004
Publisher: Elsevier BV
Date: 2015
Publisher: CRC Press
Date: 26-11-2019
Publisher: Wiley
Date: 13-07-2009
Publisher: Elsevier BV
Date: 02-2009
Publisher: Elsevier BV
Date: 03-2020
Publisher: Elsevier BV
Date: 12-2015
Publisher: Elsevier BV
Date: 05-2021
Publisher: Copernicus GmbH
Date: 18-08-2020
Abstract: Abstract. The use of complex models such as deep neural networks has yielded large improvements in predictive tasks in many fields including digital soil mapping. One of the concerns about using these models is that they are perceived as black boxes with low interpretability. In this paper we introduce the use of game theory, specifically Shapley additive explanations (SHAP) values, in order to interpret a digital soil mapping model. SHAP values represent the contribution of a covariate to the final model predictions. We applied this method to a multi-task convolutional neural network trained to predict soil organic carbon in Chile. The results show the contribution of each covariate to the model predictions in three different contexts: (a) at a local level, showing the contribution of the various covariates for a single prediction (b) a global understanding of the covariate contribution and (c) a spatial interpretation of their contributions. The latter constitutes a novel application of SHAP values and also the first detailed analysis of a model in a spatial context. The analysis of a SOC (soil organic carbon) model in Chile corroborated that the model is capturing sensible relationships between SOC and rainfall, temperature, elevation, slope, and topographic wetness index. The results agree with commonly reported relationships, highlighting environmental thresholds that coincide with significant areas within the study area. This contribution addresses the limitations of the current interpretation of models in digital soil mapping, especially in a spatial context. We believe that SHAP values are a valuable tool that should be included within the DSM (digital soil mapping) framework, since they address the important concerns regarding the interpretability of more complex models. The model interpretation is a crucial step that could lead to generating new knowledge to improve our understanding of soils.
Publisher: Elsevier BV
Date: 04-2019
Publisher: Elsevier BV
Date: 02-2020
DOI: 10.1016/J.SCITOTENV.2019.134723
Abstract: Microplastics are emerging pollutants that exist in our environment. Microplastics are synthetic polymers that have particles size smaller than 5 mm. Rapid screening of microplastics contamination in the soil could assist in identifying anomalous concentrations of microplastics in the terrestrial environment. Because there is no rule on the maximum concentration limit on how much microplastics can exist within the soil, the concentration of microplastics collected from industrial areas around metropolitan Sydney was used as a baseline. Spectra obtained from the visible-near-infrared (vis-NIR) spectra has been shown to be feasible in predicting microplastics in the soil. Instead of creating a regression model predicting the concentration of microplastic, a classification model for screening was proposed. A convolutional neural network (CNN) model was trained to classify the soil s le into various degrees of contamination based on concentration. We also delved into the CNN model to understand how the CNN model classifies the spectral data input. The model performance was first tested on two levels of classification (contaminated vs. non-contaminated). The model was able to classify the uncontaminated s les into the appropriate class more accurately than the contaminated s les. When the number of classes were gradually increased, the classification accuracy for the higher level of contaminated s les improved. Transfer learning CNN model further improved the classification prediction only on the extremes, but not the intermediate classes.
Publisher: Elsevier BV
Date: 2014
Publisher: Elsevier BV
Date: 09-2020
Publisher: Elsevier
Date: 2023
Publisher: Wiley
Date: 06-2017
DOI: 10.1111/SUM.12352
Publisher: Wiley
Date: 22-08-2021
DOI: 10.1002/CCHE.10474
Abstract: Flour millers are faced with constraints of having to meet proximate specifications, usually defined by supply contracts, while trying to maximize yield. The study investigated the application of response surface methodology (RSM), in a commercial scale flour mill, as a means of maximizing yield while meeting quality constraints. This study utilized a Box–Behnken design to develop mathematical models using RSM to describe the effect of three independent variables, wheat conditioning level (12%–18%), first break roll gap (350–600 µm), and second break roll gap (200–600 µm) on the responses. The model R 2 for the responses was .98, .96, .98, and .98 for ash, protein and moisture contents, and flour yield, respectively. All models were statistically significant ( p .05) and validated with four independent experiments. RSM models were used to optimize the process to produce flour with a protein content greater than 12.0%, ash content less than 0.54%, moisture content less than 14.5%, and yield greater than 84.2% on a clean wheat, unconditioned basis. This was achieved with conditioning wheat to 18.4%, first break roll gap of 450 µm, and second break roll gap of 250 µm. The optimum mill settings resulted in a flour yield increase of 1.45% and reduction in ash content from 0.58% to 0.53%. Using RSM, significant financial gains could be achieved by producing more flour from a given quantity of wheat with lower levels of bran contamination.
Publisher: Elsevier BV
Date: 03-2015
Start Date: 04-2005
End Date: 10-2009
Amount: $298,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 05-2016
End Date: 05-2021
Amount: $226,094.00
Funder: Australian Research Council
View Funded ActivityStart Date: 01-2008
End Date: 12-2013
Amount: $880,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2019
End Date: 12-2023
Amount: $450,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2010
End Date: 10-2014
Amount: $228,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 01-2003
End Date: 02-2004
Amount: $50,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 06-2009
End Date: 12-2014
Amount: $336,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 04-2012
End Date: 12-2015
Amount: $417,458.00
Funder: Australian Research Council
View Funded ActivityStart Date: 2013
End Date: 12-2016
Amount: $686,356.00
Funder: Australian Research Council
View Funded ActivityStart Date: 07-2011
End Date: 06-2017
Amount: $365,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 05-2014
End Date: 12-2019
Amount: $530,536.00
Funder: Australian Research Council
View Funded ActivityStart Date: 03-2004
End Date: 11-2008
Amount: $300,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 01-2020
End Date: 06-2023
Amount: $455,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 03-2007
End Date: 02-2010
Amount: $330,000.00
Funder: Australian Research Council
View Funded ActivityStart Date: 12-2004
End Date: 12-2007
Amount: $453,000.00
Funder: Australian Research Council
View Funded Activity