Predictive Soil Mapping (PSM) is based on applying statistical and/or machine learning techniques to fit models for the purpose of producing spatial and/or spatiotemporal predictions of soil variables i.e. maps of soil properties and classes at different resolutions. It is a multidisciplinary field combining statistics, data science, soil science, physical geography, remote sensing, geoinformation science and a number of other sciences. Predictive Soil Mapping with R is about understanding the main concepts behind soil mapping, mastering R packages that can be used to produce high quality soil maps, and about optimizing all processes involved so that also the production costs can be reduced. The online version of the book is available at: https: //envirometrix.github.io/PredictiveSoilMapping/ Pull requests and general comments are welcome. These materials are based on technical tutorials initially developed by the ISRIC's Global Soil Information Facilities (GSIF) development team over the period 2014-2017
This book describes and provides many detailed examples of implementing Digital Soil Mapping (DSM) using R. The work adheres to Digital Soil Mapping theory, and presents a strong focus on how to apply it. DSM exercises are also included and cover procedures for handling and manipulating soil and spatial data in R. The book also introduces the basic concepts and practices for building spatial soil prediction functions, and then ultimately producing digital soil maps.
The book compiles the main ideas and methodologies that have been proposed and tested within these last fifteen years in the field of Digital Soil Mapping (DSM). Begining with current experiences of soil information system developments in various regions of the world, this volume presents states of the art of different topics covered by DSM: Conception and handling of soil databases, sampling methods, new soil spatial covariates, Quantitative spatial modelling, Quality assessment and representation of DSM outputs. This book provides a solid support to students, researchers and engineers interested in modernising soil survey approaches with numerical techniques. It is also of great interest for potential soil data users.* A new concept to meet the worldwide demand for spatial soil data * The first compilation of ideas and methodologies of Digital Soil Mapping * Offers a variety of specialities: soil surveying, geostatistics, data mining, fuzzy logic, remote sensing techniques, Geographical Information Science,...* Written by 82 researchers from 13 different countries
The Soil Organic Carbon Mapping cookbook provides a step-by-step guidance for developing 1 km grids for soil carbon stocks. It includes the preparation of local soil data, the compilation and pre-processing of ancillary spatial data sets, upscaling methodologies, and uncertainty assessments. Guidance is mainly specific to soil carbon data, but also contains many generic sections on soil grid development, as it is relevant for other soil properties. This second edition of the cookbook provides generic methodologies and technical steps to produce SOC maps and has been updated with knowledge and practical experiences gained during the implementation process of GSOCmap V1.0 throughout 2017. Guidance is mainly specific to SOC data, but as this cookbook contains generic sections on soil grid development it can be applicable to map various soil properties.
Digital Soil Mapping is the creation and the population of a geographically referenced soil database. It is generated at a given resolution by using field and laboratory observation methods coupled with environmental data through quantitative relationships. Digital soil mapping is advancing on different fronts at different rates all across the world. This book presents the state-of-the art and explores strategies for bridging research, production, and environmental application of digital soil mapping.It includes examples from North America, South America, Europe, Asia, and Australia. The chapters address the following topics: - evaluating and using legacy soil data - exploring new environmental covariates and sampling schemes - using integrated sensors to infer soil properties or status - innovative inference systems predicting soil classes, properties, and estimating their uncertainties - using digital soil mapping and techniques for soil assessment and environmental application - protocol and capacity building for making digital soil mapping operational around the globe.
GlobalSoilMap: Basis of the global spatial soil information system contains contributions that were presented at the 1st GlobalSoilMap conference, held 7-9 October 2013 in Orléans, France. These contributions demonstrate the latest developments in the GlobalSoilMap project and digital soil mapping technology for which the ultimate aim is to produce a high resolution digital spatial soil information system of selected soil properties and their uncertainties for the entire world. GlobalSoilMap: Basis of the global spatial soil information system aims to stimulate capacity building and new incentives to develop full GlobalSoilMap products in all parts of the world.
Signi?cant technological advances have been few and far between in the past approximately one hundred years of soil survey activities. Perhaps one of the most innovative techniques in the history of soil survey was the introduction of aerial photographs as base maps for ?eld mapping, which replaced the conventional base map laboriously prepared by planetable and alidade. Such a relatively simple idea by today’s standards revolutionized soil surveys by vastly increasing the accuracy and ef?ciently. Yet, even this innovative approach did not gain universal acceptance immediately and was hampered by a lack of aerial coverage of the world, funds to cover the costs, and in some cases a reluctance by some soil mappers and cartog- phers to change. Digital Soil Mapping (DSM), which is already being used and tested by groups of dedicated and innovative pedologists, is perhaps the next great advancement in delivering soil survey information. However, like many new technologies, it too has yet to gain universal acceptance and is hampered by ignorance on the part of some pedologists and other scientists. DSM is a spatial soil information system created by numerical models that - count for the spatial and temporal variations of soil properties based on soil - formation and related environmental variables (Lagacheric and McBratney, 2007).
Geocomputation with R is for people who want to analyze, visualize and model geographic data with open source software. It is based on R, a statistical programming language that has powerful data processing, visualization, and geospatial capabilities. The book equips you with the knowledge and skills to tackle a wide range of issues manifested in geographic data, including those with scientific, societal, and environmental implications. This book will interest people from many backgrounds, especially Geographic Information Systems (GIS) users interested in applying their domain-specific knowledge in a powerful open source language for data science, and R users interested in extending their skills to handle spatial data. The book is divided into three parts: (I) Foundations, aimed at getting you up-to-speed with geographic data in R, (II) extensions, which covers advanced techniques, and (III) applications to real-world problems. The chapters cover progressively more advanced topics, with early chapters providing strong foundations on which the later chapters build. Part I describes the nature of spatial datasets in R and methods for manipulating them. It also covers geographic data import/export and transforming coordinate reference systems. Part II represents methods that build on these foundations. It covers advanced map making (including web mapping), "bridges" to GIS, sharing reproducible code, and how to do cross-validation in the presence of spatial autocorrelation. Part III applies the knowledge gained to tackle real-world problems, including representing and modeling transport systems, finding optimal locations for stores or services, and ecological modeling. Exercises at the end of each chapter give you the skills needed to tackle a range of geospatial problems. Solutions for each chapter and supplementary materials providing extended examples are available at https://geocompr.github.io/geocompkg/articles/. Dr. Robin Lovelace is a University Academic Fellow at the University of Leeds, where he has taught R for geographic research over many years, with a focus on transport systems. Dr. Jakub Nowosad is an Assistant Professor in the Department of Geoinformation at the Adam Mickiewicz University in Poznan, where his focus is on the analysis of large datasets to understand environmental processes. Dr. Jannes Muenchow is a Postdoctoral Researcher in the GIScience Department at the University of Jena, where he develops and teaches a range of geographic methods, with a focus on ecological modeling, statistical geocomputing, and predictive mapping. All three are active developers and work on a number of R packages, including stplanr, sabre, and RQGIS.
This book presents the basic concepts of quantitative soil science and, within this framework, it seeks to construct a new body of knowledge. There is a growing need for quantitative approach in soil science, which arises from a general demand for improved economic production and environmental management. Pedometrics can be defined as the development and application of statistical and mathematical methods applicable to data analysis problems in soil science. This book shows how pedometrics can address key soil-related questions from a quantitative point of view. It addresses four main areas which are akin to the problems of conventional pedology: (i) Understanding the pattern of soil distribution in character space – soil classification, (ii) Understanding soil spatial and temporal variation, (iii) Evaluating the utility and quality of soil and ultimately, (iv) Understanding the genesis of soil. This is the first book that address these problems in a coherent quantitate approach.
This book reports on developments in Proximal Soil Sensing (PSS) and high resolution digital soil mapping. PSS has become a multidisciplinary area of study that aims to develop field-based techniques for collecting information on the soil from close by, or within, the soil. Amongst others, PSS involves the use of optical, geophysical, electrochemical, mathematical and statistical methods. This volume, suitable for undergraduate course material and postgraduate research, brings together ideas and examples from those developing and using proximal sensors and high resolution digital soil maps for applications such as precision agriculture, soil contamination, archaeology, peri-urban design and high land-value applications, where there is a particular need for high spatial resolution information. The book in particular covers soil sensor sampling, proximal soil sensor development and use, sensor calibrations, prediction methods for large data sets, applications of proximal soil sensing, and high-resolution digital soil mapping. Key themes: soil sensor sampling – soil sensor calibrations – spatial prediction methods – reflectance spectroscopy – electromagnetic induction and electrical resistivity – radar and gamma radiometrics – multi-sensor platforms – high resolution digital soil mapping - applications Raphael A. Viscarra Rossel is a scientist at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) of Australia. Alex McBratney is Pro-Dean and Professor of Soil Science in the Faculty of Agriculture Food & Natural Resources at the University of Sydney in Australia. Budiman Minasny is a Senior Research Fellow in the Faculty of Agriculture Food & Natural Resources at the University of Sydney in Australia.
