This page has only limited features, please log in for full access.
Soil ecosystem services (ES) provide multiple benefits to human well-being, but the failure to appreciate them has led to soil degradation issues across the globe. Despite an increasing interest in the threats to soil resources, economic valuation in this context is limited. Importantly, most of the existing valuation studies do not account for the spatial distribution of benefits that soil ES provide to the population. In this study, we present the results of a choice experiment (CE) aimed at investigating spatial heterogeneity of attitudes and preferences towards soil conservation and soil ES. We explored spatial heterogeneity of both attitudes and welfare measures via GIS techniques. We found that citizens of the Veneto Region (Northeast Italy) generally have positive attitudes towards soil conservation. We also find positive willingness-to-pay (WTP) values for soil ES in most of the study area and a considerable degree of heterogeneity in the spatial taste distribution. Finally, our results suggest that respondents with pro-environmental attitudes display a higher WTP based on the geographic pattern of the distribution of WTP values and attitudinal scores across the area.
Luisa Eusse-Villa; Cristiano Franceschinis; Mara Thiene; Jürgen Meyerhoff; Alex McBratney; Damien Field. Attitudes and Preferences towards Soil-Based Ecosystem Services: How Do They Vary across Space? Sustainability 2021, 13, 8722 .
AMA StyleLuisa Eusse-Villa, Cristiano Franceschinis, Mara Thiene, Jürgen Meyerhoff, Alex McBratney, Damien Field. Attitudes and Preferences towards Soil-Based Ecosystem Services: How Do They Vary across Space? Sustainability. 2021; 13 (16):8722.
Chicago/Turabian StyleLuisa Eusse-Villa; Cristiano Franceschinis; Mara Thiene; Jürgen Meyerhoff; Alex McBratney; Damien Field. 2021. "Attitudes and Preferences towards Soil-Based Ecosystem Services: How Do They Vary across Space?" Sustainability 13, no. 16: 8722.
The assessment of changes in soil condition and capability requires the identification of a reference state specific to each soil class. This study develops a framework for mapping soil classes that can be used as a reference state. It identifies soil classes that should have undergone similar historic anthropedogenesis, and differentiate, within each class, zones that have been less affected by human activities. This approach could be used as a baseline for assessing contemporary soil change, as demonstrated in the state of New South Wales in Australia. First, we established soil classes with similar multimillennial natural pedogenesis and historic anthropedogenesis, called pedogenons. This was achieved by applying unsupervised classification (k-means) to a set of quantitative state variables, proxies of the soil-forming factors at the time of the European settlement in New South Wales (climate, relief, parent material, and estimated pre-1750s vegetation). Pedogenon classes were then stratified into subclasses (ranging from remnant pedogenons to different pedophenons) by combining information on native vegetation extent, status (remnant or cleared) and current land use (i.e., land use history). The stratification of 1000 pedogenon classes resulted in 5448 subclasses, ranging from remnant pedogenons (located in protected areas of intact native vegetation), quasi-remnant pedogenons (production with low intervention on remnant native vegetation), cleared, grazing, and cropping pedophenons. The median of the area proportion of the pedogenon that was still preserved as remnant vegetation was 5.3%. This quasi-remnant pedogenon or the less affected pedophenon could be used as reference state. Pedophenon grazing and cropping occupied larger areas, with mean values of 73 km2 and 153 km2 respectively. The application of this framework for assessing soil change is illustrated using legacy data of topsoil pH (5 – 15 cm) as one indicator of soil condition. The ability of the pedogenon and pedophenon subclasses for explaining the variation of three stable (total Si, total Al, clay) and three dynamic (bulk density, particulate organic carbon, pH) soil properties from agricultural soils. A generalised least squares model indicated that the effects of pedogenon, land use history and their interaction on topsoil pH were statistically significant (p < 0.001). Paired comparisons between pedogenon/pedophenon subclasses by pedogenon class were not statistically significant, although we observed the general trend: remnant pedogenon ≈ quasi-remnant pedogenon < pedophenon cleared ≈ pedophenon grazing < pedophenon cropping. Redundancy discriminant analysis indicated that pedogenons explained 40 % of the variation of stable and dynamic soil properties, pedogenon/pedophenon subclasses explained 0.1 % and the shared effect explained 18 %, leaving 42 % of unexplained variance. The effects of pedogenon/pedophenon subclasses on the location of group centroids were statistically significant only when dynamic soil properties were considered, but not for stable and dynamic soil properties. This framework can be integrated into a soil security assessment once the indicators of soil condition and capability are translated into soil functions and ecosystem services. Other potential applications include the design of soil monitoring sampling schemes and identifying thresholds of soil degradation.
Mercedes Román Dobarco; Alex McBratney; Budiman Minasny; Brendan Malone. Setting up a framework to assess changes in soil condition and capability over large areas. Soil Security 2021, 100011 .
AMA StyleMercedes Román Dobarco, Alex McBratney, Budiman Minasny, Brendan Malone. Setting up a framework to assess changes in soil condition and capability over large areas. Soil Security. 2021; ():100011.
Chicago/Turabian StyleMercedes Román Dobarco; Alex McBratney; Budiman Minasny; Brendan Malone. 2021. "Setting up a framework to assess changes in soil condition and capability over large areas." Soil Security , no. : 100011.
The response of soils to different human forcings may vary among soil classes (in magnitude and direction of change) depending on their resistance and resilience. We propose a modelling framework for mapping soil-class specific references (i.e., genosoils) and their variants (i.e., phenosoils) that can be used for assessing changes in soil condition due to land use change and management practices. The methodology consists of a first step that creates groups characterized by homogeneous soil-forming factors for a given reference time, under the hypothesis that these groups represent soil classes resulting from multimillennial natural pedogenesis and historic anthropedogenesis (i.e., soil formation processes modified by human activities) (i.e., pedogenons). In this study we applied the methodology to New South Wales (Australia) at the time of the European settlement, because from 1788 onwards the intensification of land use may have accelerated the rate of change of soil properties. A thousand pedogenon classes were generated applying k-means clustering to a set of quantitative state variables that represent the soil-forming factors at the time of the European settlement. Hierarchical clustering was applied to the centroids of the pedogenon classes for assessing their similarities and organization. In a second step, information on native vegetation extent, status (cleared or intact), and current land use was combined for creating a categorical map distinguishing areas with different expected degree of human-induced soil change. The combination of both maps resulted in 5448 subclasses, ranging from remnant genosoils (located in protected areas of intact native vegetation), genosoils II, cleared, grazing and cropping phenosoils. For each pedogenon there was at least a 90-m grid cell classified as a remnant genosoil. The median of the proportion of the pedogenon of origin preserved as a remnant genosoil was 5.3%. Phenosoils grazing and cropping occupied larger areas, with mean values of 73 km2 and 153 km2 respectively. Finally, we tested differences in topsoil pH, as proxy for soil condition, by genosoil and phenosoil classes using legacy soil data accessed with the Soil Data Federator from the Terrestrial Ecosystem Research Network. A gls model indicated that the effects of pedogenon, genosoil/phenosoil and their interaction were statistically significant (p < 0.001). Paired mean comparisons suggested that mean pH did not differ between remnant genosoils and genosoil II, but the mean pH of both genosoil classes differed from phenosoils. Estimated pH means did not differ between phenosoil classes, although it followed the trend remnant genosoil < genosoil II < phenosoil cleared < phenosoil grazing < phenosoil cropping. The proposed methodology has several potential applications, including soil security and soil change assessment, and designing soil monitoring surveys.
Mercedes Roman Dobarco; Alex McBratney; Budiman Minasny; Brendan Malone. Digital pedogenon mapping as basis for assessing changes in soil condition. 2021, 1 .
AMA StyleMercedes Roman Dobarco, Alex McBratney, Budiman Minasny, Brendan Malone. Digital pedogenon mapping as basis for assessing changes in soil condition. . 2021; ():1.
Chicago/Turabian StyleMercedes Roman Dobarco; Alex McBratney; Budiman Minasny; Brendan Malone. 2021. "Digital pedogenon mapping as basis for assessing changes in soil condition." , no. : 1.
Soil aggregate stability is a useful indicator of soil physical health and can be used to monitor condition through time. A novel method of quantifying soil aggregate stability, based on the relative increase in the footprint area of aggregates as they disintegrate when immersed in water, has been developed and can be performed using a smartphone application – SLAKES. In this study the SLAKES application was used to obtain slaking index (SI) values of topsoil samples (0 to 10 cm) at 158 sites to assess aggregate stability in a mixed agricultural landscape. A large range in SI values of 0 to 7.3 was observed. Soil properties and land use were found to be correlated with observed SI values. Soils with clay content >25 % and cation exchange capacity (CEC) : clay ratio >0.5 had the highest observed SI values. Variation in SI for these soils was driven by organic carbon (OC) content which fit a segmented exponential decay function. An OC threshold of 1.1 % was observed, below which the most extreme SI values were observed. Soils under dryland and irrigated cropping had lower OC content and higher observed SI values compared to soils under perennial cover. These results suggest that farm managers can mitigate the effects of extreme slaking by implementing management practices to increase OC content, such as minimum tillage or cover cropping. A regression-kriging method utilising a Cubist model with a suite of spatial covariates was used to map SI across the study area. Accurate predictions were produced with leave-one-out cross-validation, giving a Lin's concordance correlation coefficient (LCCC) of 0.85 and a root-mean-square error (RMSE) of 1.1. Similar validation metrics were observed in an independent test set of samples consisting of 50 observations (LCCC = 0.82; RMSE = 1.1). The potential impact of implementing management practices that promote soil OC sequestration on SI values in the study area was explored by simulating how a 0.5 and 1.0 % increase in OC would impact SI values at observation points and then mapping this across the study area. Overall, the maps produced in this study have the potential to guide management decisions by identifying areas that currently experience extreme slaking and highlighting areas that are expected to have a significant reduction in slaking by increasing OC content.
Edward J. Jones; Patrick Filippi; Rémi Wittig; Mario Fajardo; Vanessa Pino; Alex B. McBratney. Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape. SOIL 2021, 7, 33 -46.
AMA StyleEdward J. Jones, Patrick Filippi, Rémi Wittig, Mario Fajardo, Vanessa Pino, Alex B. McBratney. Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape. SOIL. 2021; 7 (1):33-46.
Chicago/Turabian StyleEdward J. Jones; Patrick Filippi; Rémi Wittig; Mario Fajardo; Vanessa Pino; Alex B. McBratney. 2021. "Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape." SOIL 7, no. 1: 33-46.
Cristine Morgan; Alex McBratney. Editorial: Widening the disciplinary study of soil. Soil Security 2020, 1, 100003 .
AMA StyleCristine Morgan, Alex McBratney. Editorial: Widening the disciplinary study of soil. Soil Security. 2020; 1 ():100003.
Chicago/Turabian StyleCristine Morgan; Alex McBratney. 2020. "Editorial: Widening the disciplinary study of soil." Soil Security 1, no. : 100003.
Soil is a complex system in which biological, chemical and physical interactions take place. The behaviour of these interactions changes in spatial scale from the atomic to the global, and in time. To understand how this system works, soil scientists usually rely on incremental improvements in the knowledge by refinement of theories through hypothesis testing and development using carefully designed experiments. In the last two decades, the primacy of this knowledge construction process has been challenged by the development of large soil databases and algorithms such as machine learning. The data-driven research approach to soil science, the inference of soil knowledge directly from data by using computational tools and modelling techniques, is becoming more popular. Despite the wide adoption of a data-driven research approach to soil science, there has been little discussion on how a research driven by data instead of hypotheses affects scientific progress. In this paper, we provide an introductory perspective on data-driven soil research by discussing some of the issues and opportunities of knowledge discovery from soil data. We show that while data-driven soil research may seem revolutionary for some, soil science has a long history of exploratory efforts to generate knowledge from data. Empirical and factual soil classifications, for example, were data driven. We further discuss, with examples, (i) data, databases and the logic of data storage for data-driven soil research, (ii) the issues of extreme empiricist claims that arise corollary to the increase in the use of computational tools, and (iii) the challenge of formulating a scientific explanation based on patterns observed in the data and data analysis tools. By considering the epistemic challenges of the data-driven scientific research in the light of the historical literature, we found that there is a continuity of practices, some being certainly amplified by recent technological changes, but that the core methods of scientific enquiry from data remain essentially unchanged. Highlights Historical account of data-driven soil science research. Describe data to be used for data-driven soil science. Discuss conceptual issues and opportunities for data-driven soil science. Investigate the challenge of formulating an explanation from soil data.
Alexandre M. J.‐C. Wadoux; Mercedes Román‐Dobarco; Alex B. McBratney. Perspectives on data‐driven soil research. European Journal of Soil Science 2020, 72, 1675 -1689.
AMA StyleAlexandre M. J.‐C. Wadoux, Mercedes Román‐Dobarco, Alex B. McBratney. Perspectives on data‐driven soil research. European Journal of Soil Science. 2020; 72 (4):1675-1689.
Chicago/Turabian StyleAlexandre M. J.‐C. Wadoux; Mercedes Román‐Dobarco; Alex B. McBratney. 2020. "Perspectives on data‐driven soil research." European Journal of Soil Science 72, no. 4: 1675-1689.
This paper provides a history of the investigation of the soils and organic matter of Deli in Sumatra, Indonesia, for growing tobacco in the early 20th century and an interpretation based on current data, knowledge and understanding. We first review some early chemists and agrogeologists’ investigations on the soils of Deli to increase tobacco production. Van Bemmelen studied the humus of the soil of Deli in 1890 and formalised an 8-year fallow plantation scheme for growing tobacco. While maintaining organic matter had been established, the complexity of soil distribution in the area was more important in determining the quality of tobacco. It took another 40 years for the soil in Deli area to be properly mapped. Jan Henri Druif in the 1930s mapped and classified the soils of Deli based on their parent material and mineralogical composition. We then describe the rise and demise of the tobacco industry from 1930s-current. We examine the implication of the fallow system and soil distribution with the current understanding of soil carbon processes and recent data. The results are interpreted and discussed considering i) the myth of ”poor” tropical soils, ii) nutrient availability after slash and burn, iii) soil organic matter decline after forest conversion and recovery after fallow, and iv) soil mapping and provenance. Based on published studies and observed data coupled with modelling, we attempt to explain early researchers’ observations and deductions. We summarise soil organic carbon dynamic conditions in the tropics after 50 years of forest clearance: under fallow rotation, it is possible to maintain, on average, a constant value of 20% organic carbon (OC) decrease from the original level, while continuous cropping can decrease OC levels up to 30–40%. An extreme condition with continuous cultivation and little organic matter input can result in an OC decline of up to 80%. The historical studies enable to appreciate aspects of soil mapping and organic matter that are repeatedly overlooked in present-day research.
Budiman Minasny; Erwin Nyak Akoeb; Tengku Sabrina; Alexandre Wadoux; Alex B. McBratney. History and interpretation of early soil and organic matter investigations in Deli, Sumatra, Indonesia. CATENA 2020, 195, 104909 .
AMA StyleBudiman Minasny, Erwin Nyak Akoeb, Tengku Sabrina, Alexandre Wadoux, Alex B. McBratney. History and interpretation of early soil and organic matter investigations in Deli, Sumatra, Indonesia. CATENA. 2020; 195 ():104909.
Chicago/Turabian StyleBudiman Minasny; Erwin Nyak Akoeb; Tengku Sabrina; Alexandre Wadoux; Alex B. McBratney. 2020. "History and interpretation of early soil and organic matter investigations in Deli, Sumatra, Indonesia." CATENA 195, no. : 104909.
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.
José Padarian; Alex B. McBratney; Budiman Minasny. Game theory interpretation of digital soil mapping convolutional neural networks. SOIL 2020, 6, 389 -397.
AMA StyleJosé Padarian, Alex B. McBratney, Budiman Minasny. Game theory interpretation of digital soil mapping convolutional neural networks. SOIL. 2020; 6 (2):389-397.
Chicago/Turabian StyleJosé Padarian; Alex B. McBratney; Budiman Minasny. 2020. "Game theory interpretation of digital soil mapping convolutional neural networks." SOIL 6, no. 2: 389-397.
Soil aggregate stability is a useful indicator of soil physical health and can be used to monitor condition through time. A novel method to quantify soil aggregate stability, based on the relative increase in the footprint area of aggregates as they disintegrate when immersed in water, has been developed and can be performed using a smartphone application – SLAKES. In this study the SLAKES application was used to obtain slaking index (SI) values of topsoil samples (0 to 10 cm) at 158 sites to assess aggregate stability in a mixed agricultural landscape. A large range in SI values of 0 to 7.3 was observed. Soil properties and land use were found to be correlated with observed SI values. Soils with clay content > 25 % and CEC : clay ratio > 0.5 had the highest observed SI values. Variation in SI for these soils was driven by OC content which fit a segmented exponential decay function. An OC threshold of 1.1 % was observed below which the most extreme SI values were observed. Soils under dryland and irrigated cropping had lower OC content and higher observed SI values compared to soils under perennial cover. These results suggest that farm managers can mitigate the effects of extreme slaking by implementing management practices to increase OC content, such as minimum tillage or cover-cropping. A regression-kriging method utilising a Cubist model with a suite of spatial covariates was used to map SI across the study area. Accurate predictions were produced with leave-one-out cross-validation (LOOCV) giving an LCCC of 0.85 and an RMSE of 1.1. Similar validation metrics were observed in an independent test set of samples consisting of 50 observations (LCCC = 0.82; RMSE = 1.1). The potential impact of implementing management practices that promote soil OC sequestration on SI values in the study area was explored by simulating how a 1 % increase in OC would impact SI values at observation points, and then mapping this across the study area. Overall, the maps produced in this study have the potential to guide management decisions by identifying areas that currently experience extreme slaking, and those areas that are expected to have a significant reduction in slaking by increasing OC content.
Edward J. Jones; Patrick Filippi; Rémi Wittig; Mario Fajardo; Vanessa Pino; Alex B. McBratney. Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape. 2020, 2020, 1 -22.
AMA StyleEdward J. Jones, Patrick Filippi, Rémi Wittig, Mario Fajardo, Vanessa Pino, Alex B. McBratney. Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape. . 2020; 2020 ():1-22.
Chicago/Turabian StyleEdward J. Jones; Patrick Filippi; Rémi Wittig; Mario Fajardo; Vanessa Pino; Alex B. McBratney. 2020. "Mapping soil slaking index and assessing the impact of management in a mixed agricultural landscape." 2020, no. : 1-22.
The use of complex models such as deep neural networks has yielded large improvements in predictive tasks in many fields including digital soil mapping. Once 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 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 of 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 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 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.
José Padarian; Alex B. McBratney; Budiman Minasny. Game theory interpretation of digital soil mapping convolutional neural networks. 2020, 2020, 1 -12.
AMA StyleJosé Padarian, Alex B. McBratney, Budiman Minasny. Game theory interpretation of digital soil mapping convolutional neural networks. . 2020; 2020 ():1-12.
Chicago/Turabian StyleJosé Padarian; Alex B. McBratney; Budiman Minasny. 2020. "Game theory interpretation of digital soil mapping convolutional neural networks." 2020, no. : 1-12.
Data sharing and collaboration are critical to solving large-scale problems. The prevailing soil data-sharing model is based on different groups sending their data to a lead party. This model is of a centralised nature and, consequently, results in the participants ceding control and governance over their data to the lead party. Here we explore the use of a distributed ledger (blockchain) to solve the aforementioned issues. We explain what a blockchain is and some of its characteristics to then describe some features of a blockchain that make it an interesting candidate for an inter-institutional database. Finally, we describe the potential use case of developing a global soil spectral library with multiple, independent international institutions constituting the network.
José Padarian; Alex B. McBratney. A new model for intra- and inter-institutional soil data sharing. SOIL 2020, 6, 89 -94.
AMA StyleJosé Padarian, Alex B. McBratney. A new model for intra- and inter-institutional soil data sharing. SOIL. 2020; 6 (1):89-94.
Chicago/Turabian StyleJosé Padarian; Alex B. McBratney. 2020. "A new model for intra- and inter-institutional soil data sharing." SOIL 6, no. 1: 89-94.
Agricultural pesticides can become persistent environmental pollutants. Among many, glyphosate (GLP) is under particular scrutiny because of its extensive use and its alleged threats to the ecosystem and human health. Here, we introduce the first global environmental contamination analysis of GLP and its metabolite, AMPA, conducted with a mechanistic dynamic model at 0.5 × 0.5° spatial resolution (about 55 km at the equator) fed with geographically-distributed agricultural quantities, soil and biogeochemical properties, and hydroclimatic variables. Our analyses reveal that about 1% of croplands worldwide (385,000 km2) are susceptible to mid to high contamination hazard and less than 0.1% has a high hazard. Hotspots found in South America, Europe, and East and South Asia were mostly correlated to widespread GLP use in pastures, soybean, and corn; diffuse contributing processes were mainly biodegradation recalcitrance and persistence, while soil residue accumulation and leaching below the root zone contributed locally to the hazard in hotspots. Hydroclimatic and soil variables were major controlling factors of contamination hotspots. The relatively low risk of environmental exposure highlighted in our work for a single active substance does not rule out a greater recognition of environmental pollution by pesticides and calls for worldwide cooperation to develop timely standards and implement regulated strategies to prevent excess global environmental pollution.
Federico Maggi; Daniele la Cecilia; Fiona H.M. Tang; Alexander McBratney. The global environmental hazard of glyphosate use. Science of The Total Environment 2020, 717, 137167 .
AMA StyleFederico Maggi, Daniele la Cecilia, Fiona H.M. Tang, Alexander McBratney. The global environmental hazard of glyphosate use. Science of The Total Environment. 2020; 717 ():137167.
Chicago/Turabian StyleFederico Maggi; Daniele la Cecilia; Fiona H.M. Tang; Alexander McBratney. 2020. "The global environmental hazard of glyphosate use." Science of The Total Environment 717, no. : 137167.
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.
José Padarian; Budiman Minasny; Alex B. McBratney. Machine learning and soil sciences: a review aided by machine learning tools. SOIL 2020, 6, 35 -52.
AMA StyleJosé Padarian, Budiman Minasny, Alex B. McBratney. Machine learning and soil sciences: a review aided by machine learning tools. SOIL. 2020; 6 (1):35-52.
Chicago/Turabian StyleJosé Padarian; Budiman Minasny; Alex B. McBratney. 2020. "Machine learning and soil sciences: a review aided by machine learning tools." SOIL 6, no. 1: 35-52.
Many initiatives try to integrate data from different parties to solve problems that could not be addressed by a sole participant. Despite the well-known benefits of collaboration, concerns of data privacy and confidentiality are still an obstacle that impedes progress in collaborative global research. This work tackles this issue using an online-learning algorithm to generate a single model where the data remains with each party and there is no need to integrate it to a single source. This approach is demonstrated in building a global soil organic carbon model based on databases of field observations held by 65 different countries. The model is trained by visiting each country, one at a time. Only knowledge and parameters of the model are transferred between countries. The results show that it is possible that the proposed approach yields a similar prediction accuracy compared with a model that is trained with all the data.
J. Padarian; B. Minasny; A.B. McBratney. Online machine learning for collaborative biophysical modelling. Environmental Modelling & Software 2019, 122, 104548 .
AMA StyleJ. Padarian, B. Minasny, A.B. McBratney. Online machine learning for collaborative biophysical modelling. Environmental Modelling & Software. 2019; 122 ():104548.
Chicago/Turabian StyleJ. Padarian; B. Minasny; A.B. McBratney. 2019. "Online machine learning for collaborative biophysical modelling." Environmental Modelling & Software 122, no. : 104548.
Data sharing and collaboration are critical to solving large scale problems. The prevailing soil data-sharing model is based on different groups sending their data to a lead party. This model is of a centralised nature and, consequently, results in the participants ceding their control and governance over their data to the lead party. Here we explore the use of a distributed ledger (blockchain) to solve the aforementioned issues. We explain what a blockchain is and some of its characteristics to then describe some features of a blockchain that makes it an interesting candidate for an inter-institutional database. Finally, we describe the potential use case of developing a global soil spectral library with multiple, independent international institutions constituting the network.
José Padarian; Alex B. McBratney. A new model for intra- and inter-institutional soil data sharing. 2019, 2019, 1 -9.
AMA StyleJosé Padarian, Alex B. McBratney. A new model for intra- and inter-institutional soil data sharing. . 2019; 2019 ():1-9.
Chicago/Turabian StyleJosé Padarian; Alex B. McBratney. 2019. "A new model for intra- and inter-institutional soil data sharing." 2019, no. : 1-9.
The application of machine learning (ML) techniques in various fields of science has increased rapidly, especially in the last ten 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 corpus. 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 publication can be grouped into 12 topics, namely remote sensing, soil organic carbon, water, contamination, methods (ensembles), erosion and parent material, methods (NN, SVM), spectroscopy, modelling (classes), crops, physical and modelling (continuous), 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.
José Padarian; Budiman Minasny; Alex B. McBratney. Machine learning and soil sciences: A review aided by machine learning tools. 2019, 2019, 1 -29.
AMA StyleJosé Padarian, Budiman Minasny, Alex B. McBratney. Machine learning and soil sciences: A review aided by machine learning tools. . 2019; 2019 ():1-29.
Chicago/Turabian StyleJosé Padarian; Budiman Minasny; Alex B. McBratney. 2019. "Machine learning and soil sciences: A review aided by machine learning tools." 2019, no. : 1-29.
Soil Security is an emerging sustainability science concept with global application for guiding integrated approaches to land management, while balancing ecosystem services, environmental, social, cultural, and economic imperatives. This discussion paper sets the scene for an Australian Soil Security framework as an example of how it might be developed for any country, defining the key issues and justification for Soil Security, as well as detailing implementation requirements and benefits; two examples of beneficial outcomes are provided in terms of facilitating decommoditization of agricultural products and the impact of urban encroachment on productive land. We highlight research gaps, where new knowledge will contribute to well-rounded approaches that reflect differing stakeholder perspectives. We also provide key nomenclature associated with a potential Soil Security framework so that future discussions may use a common language. Through this work we invite scientific and policy discourse with the aim of developing more informed responses to the myriad of competing demands placed on our soil systems.
John McLean Bennett; Alex McBratney; Damien Field; Darren Kidd; Uta Stockmann; Craig Liddicoat; Samantha Grover. Soil Security for Australia. Sustainability 2019, 11, 3416 .
AMA StyleJohn McLean Bennett, Alex McBratney, Damien Field, Darren Kidd, Uta Stockmann, Craig Liddicoat, Samantha Grover. Soil Security for Australia. Sustainability. 2019; 11 (12):3416.
Chicago/Turabian StyleJohn McLean Bennett; Alex McBratney; Damien Field; Darren Kidd; Uta Stockmann; Craig Liddicoat; Samantha Grover. 2019. "Soil Security for Australia." Sustainability 11, no. 12: 3416.
The role of soil in the existential environmental problems of declining biodiversity, climate change, water and energy security, impacting on food security has highlighted the need to link the soil functions to ecosystem services. We describe and illustrate by a limited example, the concepts and assessment of soil’s capacity measured through its capability and condition as contributors to an overall soil security framework. The framework is based on the concepts of genosoils and phenosoils. The links to other notions, such as threats to soil and soil functions are made. The framework can be potentially applied elsewhere to quantify soil changes under natural processes and human activities.
Alex. B. McBratney; Damien Field; Cristine L.S. Morgan; Jingyi Huang. On Soil Capability, Capacity, and Condition. Sustainability 2019, 11, 3350 .
AMA StyleAlex. B. McBratney, Damien Field, Cristine L.S. Morgan, Jingyi Huang. On Soil Capability, Capacity, and Condition. Sustainability. 2019; 11 (12):3350.
Chicago/Turabian StyleAlex. B. McBratney; Damien Field; Cristine L.S. Morgan; Jingyi Huang. 2019. "On Soil Capability, Capacity, and Condition." Sustainability 11, no. 12: 3350.
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 example 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 example 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.
José Padarian; Budiman Minasny; Alex B. McBratney. Using deep learning for digital soil mapping. SOIL 2019, 5, 79 -89.
AMA StyleJosé Padarian, Budiman Minasny, Alex B. McBratney. Using deep learning for digital soil mapping. SOIL. 2019; 5 (1):79-89.
Chicago/Turabian StyleJosé Padarian; Budiman Minasny; Alex B. McBratney. 2019. "Using deep learning for digital soil mapping." SOIL 5, no. 1: 79-89.
Alex McBratney. comment. 2019, 1 .
AMA StyleAlex McBratney. comment. . 2019; ():1.
Chicago/Turabian StyleAlex McBratney. 2019. "comment." , no. : 1.