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.
Highlights
Digital soil mapping (DSM) techniques are commonly used to predict a soil property at unsampled locations using measurements at a finite number of spatial locations
It is acknowledged that using covariates with coarse spatial resolution can provide satisfactory prediction (Samuel-Rosa et al, 2015). While these approaches enable us to contextualize the spatial information supplied to the regression, they rely either on heavy covariate preprocessing (Behrens et al, 2014), subjective decisions based on the resolution to which covariates must be treated as input to the model (Miller et al, 2015) or the modeller’s choice regarding neighbouring size (Behrens et al, 2010b). In light of these drawbacks, we propose the use of the convolutional neural network (CNN) as an alternative tool for mapping while explicitly accounting for local contextual information contained in covariates
We first describe the principle of an artificial neural network (ANN), which is the basis of the CNN
Summary
Digital soil mapping (DSM) techniques are commonly used to predict a soil property at unsampled locations using measurements at a finite number of spatial locations. Prediction is routinely done by exploiting the relationship between a soil property and one or several environmental covariates, which are assumed to represent soil forming factors. Examples of covariates are the digital elevation model (DEM) or its derivatives (Moore et al, 1993). The choice of covariates is governed either by their availability, preselected using a priori pedological expertise, or based on the pedological concepts whereby covariates must portray the factors of soil formation such as climate, organisms, relief, parent material and time (McBratney et al, 2018). The relation between soil property and the chosen covariates is modelled by a regression model which relates either linearly (Wadoux et al, 2018) or non-linearly (Grimm et al, 2008) sampled (point) soil properties and a vector of covariate values extracted at the same point location
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
