Estimation of wet aggregation indices using soil properties and diffuse reflectance near infrared spectroscopy: An application of classification and regression tree analysis

Waruru, Bernard K.; Ndegwa, George M.; Sila, Andrew

doi:10.1016/j.biosystemseng.2016.08.003

Cited by 17 publications

(8 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, more than one index was used to evaluate the aggregate stability. Partial least squares (SIMCA 14.1 Umetrics company, Malmö, Sweden) was used to further explore the relationship between the soil aggregate stability index and related variables [45]. The structural equation model (SmartPLS 3 GmbH, Oststeinbek, Germany) based on the partial least squares method was used to explore the relationship between different variables and the comprehensive influence of these variables on the stability of aggregates.…”

Section: Discussionmentioning

confidence: 99%

The Changes in Soil Microorganisms and Soil Chemical Properties Affect the Heterogeneity and Stability of Soil Aggregates before and after Grassland Conversion

et al. 2022

View full text Add to dashboard Cite

The conversion of grasslands to croplands is common in the agro-pastoral ecotone and brings potential risks to soil health and environmental safety. As the forming unit of soil structure, the status of soil aggregates determines soil health and is affected by multiple factors. This study investigated the changes in soil aggregate and main related factors in conversion grasslands with different managed years. Grassland conversion ages were selected as experimental treatments, which included unmanaged grassland, 3 years, 10 years, 30 years, and 50 years since grassland conversion. After grassland conversion, the proportion of large macro-aggregates with a particle size of >2 mm in the 0–10 cm soil layer decreased, small macro-aggregates with a particle size of 2–0.25 mm and micro-aggregates with a particle size of 0.25–0.053 mm increased, while aggregates with a particle size of <0.053 mm had no significant change. Soil chemical properties, most microorganisms and the soil aggregate stability indices MWD and GMD decreased at the early stage (<30 years) of the managed grasslands. After about 50 years of cultivation, soil chemical properties and microorganisms returned to equal or higher levels compared to unmanaged grasslands. However, the stability of aggregates (mean weight diameter (MWD) and geometric mean diameter (GMD)) did not recover to the initial state. MWD and GMD were positively correlated with most bacterial factors (total phospholipid fatty acids (PLFAs), bacteria, Gram-positive bacteria, Gram-negative bacteria, actinomycetes and arbuscular mycorrhizal fungi (AMF)) and some soil chemical properties (carbon, nitrogen and polysaccharides). According to the partial least square structural equation model, soil organic carbon, total nitrogen and phosphorus in the 0–10 cm soil layer explained 33.0% of the variance in MWD by influencing microorganisms. These results indicated that the stability of aggregates was directly driven by microorganisms and indirectly affected by soil organic carbon, total nitrogen and phosphorus.

show abstract

Section: Discussionmentioning

confidence: 99%

The Changes in Soil Microorganisms and Soil Chemical Properties Affect the Heterogeneity and Stability of Soil Aggregates before and after Grassland Conversion

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The main soil characteristics of soils affecting the aggregate stabilities are soil texture, clay mineralogy, SOM content, type and concentration of cations and sesquioxide contents (Le Bissonnais, 1996). However, general soil properties most closely correlated with soil AS are the contents of clay and SOM (Canasveras et al, 2010;Jozefaciuk and Czachor, 2014;Waruru et al, 2016). Clay particles are considered as cementing agents for aggregation because of their high specific surface area, cation exchange capacity, and consequently, high physical and chemical activity.…”

Section: Discussionmentioning

confidence: 99%

Effects of Land Covers on Hydro-Physical Soil Characteristics in Semi-Arid Environments: The Case of Gumushane-Siran, Turkey

Usta¹,

Yılmaz²,

Kocamanoğlu³

et al. 2018

Appl. Ecol. Env. Res.

View full text Add to dashboard Cite

Land degradation is already one of the major problems affecting the world. Land degradation by soil erosion is one of the most serious ecological problems and environmental issues in many sites of arid and semiarid regions. There has been an increasing concern regarding the semiarid highlands of Turkey that degradation of these ecosystems results in excessive soil erosion, ecosystem degradation, and loss of natural resources. This paper, therefore, explores the impact of land cover in a linkage to the soil erodibility indices and some soil properties of Koruluk (Gümüşhane-Turkey) dam watershed. Soil samples were taken from soil profiles opened in 78 sample pilots among the lands which border each other in the same altitude levels and same slopes on the andesite-basalt bedrock in the southern aspect group according to 0-10 cm and 10-20 cm depths. Sand, silt, clay, soil organic matter, dispersion ratio, erosion ratio, structural stability index, aggregate stability, clay dispersion ratio, field capacity, available water capacity, bulk density, total porosity, soil pH and electrical conductivity analyses were performed in soil samples. The results obtained about land cover were tested by variance analysis. The relationships between the selected soil characteristics were determined by correlation analysis. Degraded scotch pine forest soils were observed to be more susceptible to erosion compared to degraded oak forests and open space areas. Consequently, it was revealed that soils that developed from similar parent material in semiarid ecosystem and under all land covers were susceptible to erosion. The fact that the watershed supplies Koruluk dam built for irrigation water production makes these results quite significant. Agricultural activities are very important in the protection and continuity of the socio-economic situation of local people. The continuity of agricultural activities, in one respect, is associated with the availability of irrigation in this semi-arid region. Therefore, priority should be given to the necessary rehabilitation, planning and management studies in Koruluk dam watershed, and the results of this study should be considered.

show abstract

“…Pedotransfer functions are mainly used to determine the water retention properties of a soil at different matric potentials and sometimes the prediction is inaccurate using parametric models (Tietje & Tapkenhinrichs, 1993). Nonparametric approaches such as KNN and RF models have shown good predictive ability without being affected by the presence of outliers (Waruru et al., 2016). Our study therefore focused on comparing a multiple regression model with more thorough machine learning algorithms (MLAs), such as KNN, RF, XGBoost, and the multilayer neural network model (NeuralNetwork).…”

Section: Methodsmentioning

confidence: 99%

“…Pedotransfer functions are mainly used to determine the water retention properties of a soil at different matric potentials and sometimes the prediction is inaccurate using parametric models (Tietje & Tapkenhinrichs, 1993). Nonparametric approaches such as KNN and RF models have shown good predictive ability without being affected by the presence of outliers (Waruru et al, 2016). Our study therefore focused on comparing a multiple regression model with SSD, g cm −3 0.962 (0.16) 0.969 (0.4-1.9) Organic matter, % 5.5 (2.9) 5.0 (0.28-19.87) pH water 6.3 (0.72) 6.3 (5.0-8.5) CEC, cmol c kg −1 17.9 (5.7) 17.2 (0.51-65.4) P-M3, mg kg −1 57.9 (86.3) 42.8 (0.7-1,076) K-M3, mg kg −1 110.9 (107.2) 89.9 (0.0-1,966.5)…”

Section: Predictive Models To Estimate Soil Particle Size Fractionsmentioning

confidence: 99%

Machine learning models for predicting soil particle size fractions from routine soil analyses in Quebec

Martinelli

Gasser

2022

Soil Science Soc of Amer J

View full text Add to dashboard Cite

Soil texture and particle size distribution are important soil properties for understanding and interpreting the multiple processes and interactions in soil agrosystems. However, soil particle size analysis is rarely included in routine soil laboratory analyses due to cost but can be derived from other soil physico‐chemical properties and from more powerful regressions using machine learning techniques. The performance of multiple linear regression and four machine learning algorithms (MLAs)—K‐nearest neighbor (KNN), Random Forest, extra‐gradient boosting (XGBoost), and multilayer neural network (NeuralNetwork)—was compared to predict particle size fractions of clay, sand, and silt using routine analyses of soil pH, cation exchange capacity, Mehlich‐3 extracted elements, and density of dried sieved soil from 8,364 soil samples distributed across Quebec. Particle size fractions were predicted as compositional data using isometric log ratio transformation. The XGBoost model performed best, with RMSE values of 7, 10, and 12% and R2 values of .77, .57, and .73 for the prediction of clay, silt, and sand fractions, respectively. Feature importance classification varied from one model to another, but the best predictors remained the same regardless of the model used. Sieved soil density measured in the laboratory and Mehlich‐3 Mg, K, Fe, and Zn ranked as better predictors of particle size fractions. Predicting soil particle sizes using routine soil physico‐chemical properties and MLAs appears an option for incomplete legacy datasets and a promising economic alternative to current methods carried out in commercial laboratories.

show abstract

Estimation of wet aggregation indices using soil properties and diffuse reflectance near infrared spectroscopy: An application of classification and regression tree analysis

Cited by 17 publications

References 41 publications

The Changes in Soil Microorganisms and Soil Chemical Properties Affect the Heterogeneity and Stability of Soil Aggregates before and after Grassland Conversion

The Changes in Soil Microorganisms and Soil Chemical Properties Affect the Heterogeneity and Stability of Soil Aggregates before and after Grassland Conversion

Effects of Land Covers on Hydro-Physical Soil Characteristics in Semi-Arid Environments: The Case of Gumushane-Siran, Turkey

Machine learning models for predicting soil particle size fractions from routine soil analyses in Quebec

Contact Info

Product

Resources

About