Spatial Modelling and Prediction Assessment of Soil Iron Using Kriging Interpolation with pH as Auxiliary Information

Tziachris, Panagiotis; Metaxa, Eirini; Papadopoulos, Frantzis; Papadopoulou, Maria

doi:10.3390/ijgi6090283

Cited by 40 publications

(17 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The covariate importance assessments ( Figure 6) revealed that iron (Fe), manganese (Mn), and magnesium (Mg) were the most influential variables for both RF and GB. For iron (Fe) specifically, high influence on soil pH was expected according to the relevant literature [18,24]. Among the environmental variables, altitude was the one with the higher score for both ML models.…”

Section: Performance Assessmentmentioning

confidence: 99%

“…It controls soil fertility, regulates soil biogeochemical processes, and influences the structure and functioning of terrestrial ecosystems [22]. Regarding its spatial distribution, soil pH seems to be affected by terrain elevation [23] and exhibits spatial cross correlation with other elements, like soil Fe [24]. Based on the above characteristics, soil pH was considered the ideal soil parameter for the current study.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparing Machine Learning Models and Hybrid Geostatistical Methods Using Environmental and Soil Covariates for Soil pH Prediction

Tziachris¹,

Aschonitis²,

Chatzistathis³

et al. 2020

IJGI

Self Cite

View full text Add to dashboard Cite

In the current paper we assess different machine learning (ML) models and hybrid geostatistical methods in the prediction of soil pH using digital elevation model derivates (environmental covariates) and co-located soil parameters (soil covariates). The study was located in the area of Grevena, Greece, where 266 disturbed soil samples were collected from randomly selected locations and analyzed in the laboratory of the Soil and Water Resources Institute. The different models that were assessed were random forests (RF), random forests kriging (RFK), gradient boosting (GB), gradient boosting kriging (GBK), neural networks (NN), and neural networks kriging (NNK) and finally, multiple linear regression (MLR), ordinary kriging (OK), and regression kriging (RK) that although they are not ML models, they were used for comparison reasons. Both the GB and RF models presented the best results in the study, with NN a close second. The introduction of OK to the ML models’ residuals did not have a major impact. Classical geostatistical or hybrid geostatistical methods without ML (OK, MLR, and RK) exhibited worse prediction accuracy compared to the models that included ML. Furthermore, different implementations (methods and packages) of the same ML models were also assessed. Regarding RF and GB, the different implementations that were applied (ranger-ranger, randomForest-rf, xgboost-xgbTree, xgboost-xgbDART) led to similar results, whereas in NN, the differences between the implementations used (nnet-nnet and nnet-avNNet) were more distinct. Finally, ML models tuned through a random search optimization method were compared with the same ML models with their default values. The results showed that the predictions were improved by the optimization process only where the ML algorithms demanded a large number of hyperparameters that needed tuning and there was a significant difference between the default values and the optimized ones, like in the case of GB and NN, but not in RF. In general, the current study concluded that although RF and GB presented approximately the same prediction accuracy, RF had more consistent results, regardless of different packages, different hyperparameter selection methods, or even the inclusion of OK in the ML models’ residuals.

show abstract

Section: Performance Assessmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Comparing Machine Learning Models and Hybrid Geostatistical Methods Using Environmental and Soil Covariates for Soil pH Prediction

Tziachris¹,

Aschonitis²,

Chatzistathis³

et al. 2020

IJGI

Self Cite

View full text Add to dashboard Cite

show abstract

“…The UKRG method uses the assumption that the spatial variation of z is dependent on three components: a data set structure, a correlated random component, and a residual error. This is a hybrid method, where the spatial trend is measured by polynomial surfaces of different orders, derived globally or locally [45]:…”

Section: Simple Kriging (Krg)mentioning

confidence: 99%

Testing Different Interpolation Methods Based on Single Beam Echosounder River Surveying. Case Study: Siret River

Arseni

Voiculescu

Georgescu

et al. 2019

IJGI

View full text Add to dashboard Cite

Bathymetric measurements play an important role in assessing the sedimentation rate, deposition of pollutants, erosion rate, or monitoring of morphological changes in a river, lake, or accumulation basin. In order to create a coherent and continuous digital elevation model (DEM) of a river bed, various data interpolation methods are used, especially when single-beam bathymetric measurements do not cover the entire area and when there are areas which are not measured. Interpolation methods are based on numerical models applied to natural landscapes (e.g., meandering river) by taking into account various morphometric and morphologies and a wide range of scales. Obviously, each interpolation method, used in standard or customised form, yields different results. This study aims at testing four interpolation methods in order to determine the most appropriate method which will give an accurate description of the riverbed, based on single-beam bathymetric measurements. The four interpolation methods selected in the present research are: inverse distance weighting (IDW), radial basis function (RBF) with completely regularized spline (CRS) which uses deterministic interpolation, simple kriging (KRG) which is a geo-statistical method, and Topo to Raster (TopoR), a particular method specifically designed for creating continuous surfaces from various elevation points, contour, or polygon data, suitable for creating surfaces for hydrologic analysis. Digital elevation models (DEM’s) were statistically analyzed and precision and errors were evaluated. The single-beam bathymetric measurements were made on the Siret River, between 0 and 35 km. To check and validate the methods, the experiment was repeated for five randomly selected cross-sections in a 1500 m section of the river. The results were then compared with the data extracted from each elevation model generated with each of the four interpolation methods. Our results show that: 1) TopoR is the most accurate technique, and 2) the two deterministic methods give large errors in bank areas, for the entire river channel and for the particular cross-sections.

show abstract

“…In applied geostatistics, experimental variograms are approximated by the theoretical variogram models [22]. Some commonly used theoretical models mainly include the exponential, the spherical, and the Gaussian models [23].…”

Section: N)mentioning

confidence: 99%

Application of Kriging Algorithm Based on ACFPSO in Geomagnetic Data Interpolation

Zhou

Zhang

Wang

et al. 2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

High-precision geomagnetic field model is the key to magnetic anomaly detection and localization technology. The model is usually constructed through Kriging interpolation. Aiming at the problem of insufficient fitting of variogram in the existing Kriging interpolation methods, this paper proposes a particle swarm optimization algorithm with an adaptive compression factor (ACFPSO). The algorithm utilizes the degree of particle aggregation and the number of iterations to dynamically change the compression factor so as to achieve an effective balance between global optimization and local exploration. The cross-validation results show that the ACFPSO algorithm has the same convergence speed as the conventional particle swarm optimization algorithm, but the convergence accuracy is higher. Compared with the commonly used high-efficiency interpolation methods, such as the plain Kriging, the inverse distance weighting, and the radial basis function, the ACFPSO-optimized Kriging method achieves better performance (the mean absolute error is around 0.3 nT).

show abstract

Spatial Modelling and Prediction Assessment of Soil Iron Using Kriging Interpolation with pH as Auxiliary Information

Cited by 40 publications

References 31 publications

Comparing Machine Learning Models and Hybrid Geostatistical Methods Using Environmental and Soil Covariates for Soil pH Prediction

Comparing Machine Learning Models and Hybrid Geostatistical Methods Using Environmental and Soil Covariates for Soil pH Prediction

Testing Different Interpolation Methods Based on Single Beam Echosounder River Surveying. Case Study: Siret River

Application of Kriging Algorithm Based on ACFPSO in Geomagnetic Data Interpolation

Contact Info

Product

Resources

About