Variograms of Soil Properties for Agricultural and Environmental Applications

Paterson, Stacey; McBratney, Alex B.; Minasny, Budiman; Pringle, M. J.

doi:10.1007/978-3-319-63439-5_21

Cited by 8 publications

(9 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar influence distance was obtained by Jian-Bing et al (2006) and Sun et al (2003) in two grassland sites in Germany. A similar spatial correlation range was reported for croplands in an review by Paterson et al (2018), where, based on 41 variograms, the authors estimated an average spatial correlation range of around 400 m. Figure 5. Effect of vicinity size on prediction error, by depth range.…”

Section: Vicinity Sizesupporting

confidence: 79%

Using deep learning for Digital Soil Mapping

Padarian¹,

Minasny²,

McBratney³

2018

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. Digital soil mapping has been widely used as a cost-effective method for generating soil maps. However, current DSM data representation rarely incorporate 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 model that incorporates contextual information surrounding an observation to significantly improve the prediction accuracy over conventional DSM models. We describe a convolutional neural network (CNN) model that takes inputs 5 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 3D stack of images, data augmentation to reduce overfitting, and simultaneously predicting 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 the CNN model reduced the error by 30% compared with conventional techniques that only used point information of covariates. In the 10 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 covariate represented as images, it offers a simple and effective framework for future DSM models.

show abstract

Section: Vicinity Sizesupporting

confidence: 79%

Using deep learning for Digital Soil Mapping

Padarian¹,

Minasny²,

McBratney³

2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The results of the second scenario suggested that databases of variogram parameters (e.g. the one of Paterson et al (2018)) can be used to derive an average variogram, and that the latter can be used to guide sampling (McBratney & Pringle, 1999) or to predict a soil property from fewer units than usually required for estimating variogram parameters (Kerry & Oliver, 2004). An average variogram could also provide prior information for expert or Bayesian elicitation of the variogram (Cui, Stein, & Myers, 1995).…”

Section: Discussionmentioning

confidence: 99%

Efficient sampling for geostatistical surveys

Wadoux

Marchant

Lark

2019

European J Soil Science

View full text Add to dashboard Cite

A geostatistical survey for soil requires rational choices regarding the sampling strategy. If the variogram of the property of interest is known then it is possible to optimize the sampling scheme such that an objective function related to the survey error is minimized. However, the variogram is rarely known prior to sampling. Instead it must be approximated by using either a variogram estimated from a reconnaissance survey or a variogram estimated for the same soil property in similar conditions. For this reason, spatial coverage schemes are often preferred, because they rely on the simple dispersion of sampling units as uniformly as possible, and are similar to those produced by minimizing the kriging variance. If extra sampling locations are added close to those in a spatial coverage scheme then the scheme might be broadly similar to one produced by minimizing the total error (i.e. kriging variance plus the prediction error due to uncertainty in the covariance parameters). We consider the relative merits of these different sampling approaches by comparing their mean total error for different specified random functions. Our results showed the considerable benefit of adding close‐pairs to a spatial coverage scheme, and that optimizing with respect to the total error generally gave a small further advantage. When we consider the example of sampling for geostatistical survey of clay content of the soil, an optimized scheme based on the average of previously reported clay variograms was fairly robust compared to the spatial coverage plus close‐pairs scheme. We conclude that the direct optimization of spatial surveys was only rarely worthwhile. For most cases, it is best to apply a spatial coverage scheme with a proportion of additional sampling locations to provide some closely spaced pairs. Furthermore, our results indicated that the number of observations required for an effective geostatistical survey depend on the variogram parameters. Highlights We compared spatial coverage and spatial coverage plus a subset of 10% from the total sample as close‐pairs. The objective function encompasses variogram uncertainty and prediction error variance. Spatial coverage schemes always performed poorly because of the lack of information at short distances. Using a scheme in which 10% of the sampling units are taken at short distances is a robust strategy.

show abstract

“…The results of the second scenario suggested that databases of variogram parameters (e.g. the one of Paterson et al, 2018) can be used to derive an average variogram, and that the latter can be used to guide sampling (McBratney and Pringle, 1999) or to predict a soil property from fewer units than usually required for estimating variogram parameters (Kerry and Oliver, 2004). An average variogram could also provide prior information for expert or Bayesian elicitation of the variogram (Cui et al, 1995).…”

Section: Discussionmentioning

confidence: 99%

“…Here we used data from a published study on eld-scale variability of soil variograms. We used a compilation of soil clay variograms, provided by Paterson et al (2018). They were gathered from the existing literature, based on untransformed data and physical measurements.…”

Section: Scenariomentioning

confidence: 99%

“…In our second scenario we considered a geostatistical survey of soil clay content and the e ect of using the average variogram of a set presented by Paterson et al (2018) as a basis for a sampling scheme. We minimized the total prediction error variance given a sample size based on the average variogram and then repeated the tests conducted in the rst scenario to nd the size of the sc and sc + schemes that would be required to achieve the same total error as the optimized scheme for each of the clay variograms.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sampling design optimization for geostatistical modelling and prediction

Wadoux

View full text Add to dashboard Cite

Variograms of Soil Properties for Agricultural and Environmental Applications

Cited by 8 publications

References 51 publications

Using deep learning for Digital Soil Mapping

Using deep learning for Digital Soil Mapping

Efficient sampling for geostatistical surveys

Sampling design optimization for geostatistical modelling and prediction

Contact Info

Product

Resources

About