Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Mar 13;7(3):e06480.
doi: 10.1016/j.heliyon.2021.e06480. eCollection 2021 Mar.

Machine learning approaches for the prediction of soil aggregate stability

Yassine Bouslihim  1 , Aicha Rochdi  1 , Namira El Amrani Paaza  1
Affiliations

Machine learning approaches for the prediction of soil aggregate stability

Yassine Bouslihim et al. Heliyon. .

Abstract

Currently, many Pedotransfer Functions (PTFs) are being developed to predict certain soil properties worldwide, especially for difficult and time-consuming parameters to measure. However, very few studies have been done to assess the feasibility of using PTFs (regression or machine learning methods) for predicting soil aggregate stability. Also, the Random Forest (RF) method has never been used before to predict this parameter, and no study was found concerning the use of PTFs methods to estimate soil parameters in Morocco. Therefore, the current study was conducted in the three watersheds of Settat- Ben Ahmed Plateau, located in the center of Morocco and covering approximately 1000 km2. The purpose of this study is to compare the capabilities of the machine learning technique (Random Forest) and Multiple Linear Regression (MLR) to predict the Mean Weight Diameter (MWD) as an index of soil aggregate stability using soil properties from two sources data sets and remote sensing data. The performance of the models was evaluated using a 10-fold cross-validation procedure. The results achieved were acceptable in predicting soil aggregate stability and similar for both models. Thus, the addition of remote sensing indices to soil properties does not improve models. Results also show that organic matter is the most relevant variable for predicting soil aggregate stability for both models. The developed models can be used to predict the soil aggregate stability in this region and avoid waste of time and money deployed for analyses. However, we recommend using the largest and most uniform possible data set to achieve more accurate results.

Keywords: Mean weight diameter; Multiple linear regression; Pedotransfer functions; Random forest; Remote sensing data; Soil aggregate stability.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Location of the study area in the region of Chaouia Ourdigha (Settat-Ben Ahmed plateau, Morocco).
Figure 2
Figure 2
Soil classes map of A: Mazer and El Himer watersheds and B; Tamedroust watershed.
Figure 3
Figure 3
Location of the sampling points.
Figure 4
Figure 4
Soil input data used for the development of different models. (SP1: Soil Properties for 77 samples; SP2: Soil Properties for 114 samples; SPRS1: Soil Properties & Remote Sensing for 77 samples; SPRS2: Soil Properties Remote Sensing for 114 samples; LAI: Leaf Area Index; EVI: Enhanced Vegetation Index; GSI: Grain Size Index; SAVI: Soil Adjusted Vegetation Index; GVI: Green Vegetation Index; BI: Brightness Index; RI: Redness Index; SI: Salinity Index; NDWI: Normalized Difference Water Index; MSI: Moisture Stress Index; RVI: Ratio Vegetation Index; DVI: Difference Vegetation Index; NDVI: Normalized Difference Vegetation Index; TNDVI: Transformed Normalized Difference Vegetation Index).
Figure 5
Figure 5
Box plots of different soil properties for the 114 soil samples. (pH: potential of hydrogen, OM: organic matter, BD: bulk density, CEC: cation exchange capacity, AWC: available Water capacity and MWD: mean weight diameter).
Figure 6
Figure 6
Distribution of soil samples (n = 117) inside the USDA soil texture triangle (Blue: SP1 data set 77 samples, Green: BR08 data set 37samples).
Figure 7
Figure 7
Distribution of Mean Weight Diameter (MWD) for 77 samples under (fast wetting: fw, slow wetting: sw, mechanical breakdown: mb, and the mean of the three tests: MWDmean) and MWD for the 37 samples (MWDmeanBR08).
Figure 8
Figure 8
Correlation matrix between variables of different data sets. (SP1: Soil Properties for 77 samples; SP2: Soil Properties for 114 samples; SPRS1: Soil Properties & Remote Sensing indices for 77 samples and SPRS2: Soil Properties & Remote Sensing indices for 114 samples).
Figure 9
Figure 9
Variable importance rankings of the four Random Forest model. (% IncMSE: percent increase in Mean Square Error; SP1: Soil Properties for 77 samples; SP2: Soil Properties for 114 samples; SPRS1: Soil Properties & Remote Sensing for 77 samples and SPRS2: Soil Properties Remote Sensing for 114 samples).
Figure 10
Figure 10
Spatial distribution of soil aggregate stability.
Figure 11
Figure 11
Spatial distribution of A: Sand (%), B: Clay (%), C: organic matter (%) and D: soil erosion rates (t/ha/year).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Al Masmoudi Y., Bouslihim Y., Doumali K., El Aissaoui A., Ibno Namr K. Application of the random forest model to predict the plasticity state of vertisols. J. Ecol. Eng. 2021;22(2):36–46.
    1. Akpa S.I., Odeh I.O., Bishop T.F., Hartemink A.E. Digital mapping of soil particle-size fractions for Nigeria. Soil Sci. Soc. Am. J. 2014;78(6):1953–1966.
    1. Amézketa E. Soil aggregate stability: a review. J. Sustain. Agric. 1999;14(2-3):83–151.
    1. Annabi M., Raclot D., Bahri H., Bailly J.S., Gomez C., Bissonnais Y.L. Spatial variability of soil aggregate stability at the scale of an agricultural region in Tunisia. Catena. 2017;153:157–167.
    1. Anysz H., Brzozowski Ł., Kretowicz W., Narloch P. Feature importance of stabilised rammed earth components affecting the compressive strength calculated with explainable artificial intelligence tools. Materials. 2020;13(10):2317. - PMC - PubMed

LinkOut - more resources