Integrating Different Types of Knowledge for Digital Soil Mapping

Shi, Xun; Long, Robert; DeKett, Roger; Philippe, Jessica

doi:10.2136/sssaj2007.0158

Cited by 68 publications

(42 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, this might be addressed using a combination of cLHS 492 and targeted sampling or case-based reasoning (Shi et al, 2009), where the soil surveyor could manually 493 identify likely locations of rare soil types. This may be especially useful in arid and semi-arid regions 494 where small, localized areas often contain significant diversity when compared to the majority 495…”

Section: and 10) 403mentioning

confidence: 99%

Machine learning for predicting soil classes in three semi-arid landscapes

et al. 2015

View full text Add to dashboard Cite

any, machine learning model and covariate set might be optimal for predicting soil classes across 23 different landscapes. 24Our objective was to compare multiple machine learning models and covariate sets for predicting soil 25 taxonomic classes at three geographically distinct areas in the semi-arid western United States of 26 America (southern New Mexico, southwestern Utah, and northeastern Wyoming). All three areas were 27 the focus of digital soil mapping studies. Sampling sites at each study area were selected using 28 conditioned Latin hypercube sampling (cLHS). We compared models that had been used in other DSM 29 studies, including clustering algorithms, discriminant analysis, multinomial logistic regression, neural 30 networks, tree based methods, and support vector machine classifiers. Tested machine learning models 31 were divided into three groups based on model complexity: simple, moderate, and complex. We also 32

show abstract

Section: and 10) 403mentioning

confidence: 99%

Machine learning for predicting soil classes in three semi-arid landscapes

et al. 2015

View full text Add to dashboard Cite

show abstract

“…The steps required to predict solum depth were, for the most part, implemented using ArcSIE (Soil Inference Engine) version 9.2.402, a toolbox that functions as an ArcMap extension (Shi et al, 2009). ArcSIE was designed for creating soil maps using fuzzy logic and supports the knowledge-based approach to establishing relationships between the soil and its environment, providing tools for soil scientists to formalize the relationship based on pedological knowledge of the local soils.…”

Section: Knowledge-based Digital Soil Mapping Approachmentioning

confidence: 99%

“…ArcSIE provides different types of knowledge integration. In this study, rule-based reasoning was applied allowing for the covering of the entire mapping area (Shi et al, 2009).…”

Section: Knowledge-based Digital Soil Mapping Approachmentioning

confidence: 99%

“…The terrain attribute values and ranges associated with each soil map class were used to define membership or optimality functions (curves), which, in turn, define the relationship between the values of an environmental feature and soil type. The initial output from the inference is a series of fuzzy membership maps in raster format, one for each soil type under consideration (Shi et al, 2009). The fuzzy membership values represent the similarities of each pixel in the landscape to the soil types.…”

Section: Knowledge-based Digital Soil Mapping Approachmentioning

confidence: 99%

“…Such models are divided into data-driven (Pedometric approach) and knowledge-driven (Shi et al, 2009). The pedometric approach (statistic and geostatistic) gives a predictive accuracy that is generally related to a dense sampling scheme, which is not always feasible due to cost and time constraints (Zhu and Lin, 2010).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Solum depth spatial prediction comparing conventional with knowledge-based digital soil mapping approaches

Menezes

Silva

Mello

et al. 2014

Sci. agric. (Piracicaba, Braz.)

View full text Add to dashboard Cite

Solum depth and its spatial distribution play an important role in different types of environmental studies. Several approaches have been used for fitting quantitative relationships between soil properties and their environment in order to predict them spatially. This work aimed to present the steps required for solum depth spatial prediction from knowledge-based digital soil mapping, comparing the prediction to the conventional soil mapping approach through field validation, in a watershed located at Mantiqueira Range region, in the state of Minas Gerais, Brazil. Conventional soil mapping had aerial photo-interpretation as a basis. The knowledgebased digital soil mapping applied fuzzy logic and similarity vectors in an expert system. The knowledge-based digital soil mapping approach showed the advantages over the conventional soil mapping approach by applying the field expert-knowledge in order to enhance the quality of final results, predicting solum depth with suited accuracy in a continuous way, making the soillandscape relationship explicit.

show abstract