Wetting and dryning cycle effects on internal stress, particle rearrangement and soil aggregate stability

Get Complete Project Material File(s) Now! »

Aggregate stability

Aggregate stability was measured using a slightly-modified version of Le Bissonnais’ method (Le Bissonnais, 1996; ISO/DIS 10930, 2012). Specifically, air-dried samples were cut into 2-5 mm fragments.
The three stability tests of Le Bissonnais (1996) (fast wetting, slow wetting and stirring) were designed to reproduce the processes involved in crust formation and interrill erosion (slaking, differential clay swelling and mechanical breakdown). The 5 g sub-samples were dried at 40°C for 24 h before application of a test, and eac h test was replicated twice. After the tests, the resulting fragments were sieved in ethanol. The results are presented using the mean weighted diameter (MWD). Each MWD value corresponds to one of five classes of stability: MWD above 2 mm corresponds to very stable material (very low erodibility), between 2 and 1.3 mm corresponds to stable material (low erodibility), between 1.3 and 0.8 mm corresponds to median stability (median erodibility), between 0.8 and 0.4 mm corresponds to unstable material (high erodibility), and lower than 0.4 mm corresponds to very low stability (very high erodibility) (Le Bissonnais, 1996).

Standard soil properties

The standard soil properties were measured to explain the stability differences between the sites and between the crust and under-crust. The soil properties that are assessed on a regular basis by soil scientists and known to be related to aggregate stability were measured: soil texture (using laser diffraction granulometry, Mastersizer 2000, Malvern Instruments Ltd.), soil organic matter content (Walkey & Black, 1934), cation-exchange capacity (CEC) (Ammonium rapid method; Mackenzie, 1951) and pH (1:2.5 soil:water ratio, using a pH meter).

Comparison of aggregate stability for paired crust and under-crust samples

For most of the paired samples, the aggregate stability of the crust was larger than the aggregate stability of the corresponding under-crust, and the under-crust samples were never more stable than the corresponding crust (Figure 1). A correlation analysis was performed to study the relationships between the MWD of the crust and the MWD of the under-crust material. An analysis was performed for each aggregate stability test and for the mean of the three tests. The highest correlation coefficient (r=0.69) was found between the MWD of the crust and the MWD of the under-crust for the slow wetting test. For the other tests, the correlation coefficients were 0.43 (fast wetting test), 0.48 (stirring test) and 0.59 (mean of the three tests). However, these correlation coefficients were greatly influenced by the very low MWD of site C, hence making site C a hot spot. Without site C, the correlation coefficients were only 0.52 (slow wetting), 0.20 (fast wetting), -0.06 (stirring) and 0.28 (mean of the three tests).
While the difference in aggregate stability between a crust and its under-crust was always in the same direction, the amplitude of this difference varied greatly both for a given site and among the sites. For example, for the mean of the three tests, the inter-site coefficient of variation was 0.60 (Table 3a), while the intra-site coefficients of variation ranged from 0.16 (site D4) to 0.90 (site D1) (Table 3b).

Relationship between standard soil properties and aggregate stability

A correlation analysis was performed between the aggregate stability (MWD) and the standard soil properties taken as potential explanatory factors (Table 3). This analysis was performed on the crusts (Table 4a) and under-crusts (Table 4b) separately. In both cases, the highest correlation coefficients were found between the MWD of the slow wetting test and the organic matter content (0.57 and 0.56, respectively). In all cases, the CEC was significantly correlated with the MWD. For the crusts (Table 4a), pH was not significantly correlated with any of the MWD. For the under-crusts (Table 4b), pH was positively correlated with the MWD of the slow wetting test, the stirring test and the mean of the 3 stability tests. Finally, both for the crusts and under-crusts, the clay, silt and sand contents were not significantly correlated with any of the MWD.
A step-up multiple regression analysis was performed using the standard soil properties found to be significantly correlated to aggregate stability: organic matter content, CEC, and pH. For the crust, among all the tested combinations, the best regression was found for the mean MWD of the three tests as the dependent variable and for the organic matter content and CEC as the explanatory variables (moreover, the organic matter content and CEC are not independent from one another): MWDmean(mm) = 0.39(±0.15) × OM(%) + 0.06 (±0.02) × CEC – 0.66(±0.47). The coefficient o f determination (r²) was 0.38, meaning that this model explained 38% of the variation of the mean of the three aggregate stability tests. The residual standard error for the estimated MWD was 0.36 mm at the 95% confidence interval.

Relationship between standard soil properties and the difference in aggregate stability between crust and under-crust

A linear correlation analysis was performed to further attempt to link the differences in MWD between the crust and the under-crust materials to the standard soil properties (Table 5). Potential explanatory factors were the standard soil properties (as before) but also the difference between the crust and the under-crust for a given soil property. The differences in stability between the crust and the under-crust materials for the fast wetting test were positively correlated with (a) the crust organic matter content, (b) the crust and the under-crust CEC, and (c) the crust sand content but were negatively correlated with the crust silt content. For the slow wetting test, the MWD difference between the crust and under-crust materials was positively correlated with the differences in organic matter and sand content between the crust and under-crust but negatively correlated with the difference in clay content between the crust and under-crust. For the stirring test, the MWD difference between the crust and the under-crust materials was positively correlated with the crust organic matter, the crust CEC and the crust sand content but negatively correlated with the crust silt content. Considering the mean of the three stability tests, the MWD difference between the crust and the under-crust materials was positively correlated with the crust organic matter, the sand content and the difference in organic matter between the crust and under-crust but negatively correlated with the crust silt content.
A step-up multiple regression analysis was performed using the difference in aggregate stability between the crust and under-crust materials as the dependent variable and the standard soil properties and the differences between each property for the crust and under-crust as the explanatory variables. No statistically meaningful relationship was found.

Table of contents :

Chapitre 1 : Evaluation de l’érodibilité pour un sol encroûté et conséquences pour la modélisation de l’érosion
Erodibility of a crusted soil : asessment of controlling factors and consequences for erosion modelling. An example from the Loess Plateau of China
1. Abstract
2. Résumé
3. Introduction
4. Material and methods
4.1. Sampling sites
4.2. Sampling method
4.3. Measurements
4.3.1. Aggregate stability
4.3.2. Standard soil properties
4.4. Statistical analysis
5. Results
5.1. Variability of the aggregate stability
5.2. Comparison of aggregate stability for paired crust and under-crust samples
5.3. Variability of standard soil properties
5.4. Relationship between standard soil properties and aggregate stability
5.5. Relationship between standard soil properties and the difference in aggregate stability between crust and under-crust
6. Discussion
6.1. The aggregate stability of a crust is different from the aggregate stability of its undercrust material
6.2. Standard soil properties do not adequately predict aggregate stability
6.3. Consequences for erodibility assessment and erosion modeling
7. Conclusions
8. Acknowledgments
Chapitre 2 : Synthèse bibliographique des processus physico-chimiques affectant la stabilité structurale
Physico-chemical processes affecting soil aggregate stability : a review
Abstract
1. Introduction
1.1. Soil structure and aggregate stability
1.2. Factors and processes of aggregate stability variation
2. Physical and chemical processes involve in aggregate stability
2.1. Macro-aggregate scale
2.1.1. Slaking
2.1.2. Raindrop impact
2.1.3. Freezing- thawing
2.1.4. Differential swelling of clay
2.2. Micro-aggregate scale
2.2.1. Particle rearrangement during wetting drying cycles
2.2.2. Interlocking: frictional effect
2.2.3. Clay dispersion and flocculation
2.2.4. Dissolution and crystallisation
2.2.5. Age Hardening & thixotropy
2.3. Interactions between the processes during wetting and drying cycles
3. Summary of current knowledge and orientation of further investigations
3.1. Widely studied processes
3.2. Processes needing further researches
4. Conclusion
Synthèse et conclusion
Deuxième partie – Étude de terrain de la variabilité de la stabilité
structurale à pas de temps court. Évaluation de facteurs explicatifs.
Introduction
Chapitre 3 : Mesure de la variation de la stabilité structurale à pas de temps court. Conséquences pour l’estimation de l’érodibilité
Short term dynamics of aggregate stability in the field : consequences for erodibility assessment
Abstract
1. Introduction
2. Materials and Methods
2.1. Sampling sites
2.2. Monitoring setup
2.3. Sampling setup
2.4. Measurements
2.5. Statistical analysis
3. Results
3.1. Temporal variation of aggregate stability
3.1.1. Monthly variation of aggregate stability
3.1.2. Short time step variation of aggregate stability
3.2. Comparisons between aggregate stability values for the different treatments .
3.3. Relationships between aggregate stability variations for the different treatments
3.3.1. Relationship between MWD of the surface and MWD of the subsurface
3.3.2. Relationship between aggregate stability of two plots on the same site (upslope and downslope).
3.3.3. Relationship between aggregate stability of the Marcheville site and the La Gouëthière site
3.4. Relationship between aggregate stability and rain height
4. Discussion
4.1. Aggregate stability varied at short time step
4.2. Influences of the different treatments on aggregate stability temporal variability .
4.2.1. Difference between surface and subsurface samples aggregate stability
4.2.2. Difference between aggregate stability of two plots located on the same crop field
4.2.3. Differences between the aggregate stability of two field sites located in similar soil types
4.3. Relationship between aggregate stability variation and precipitation
4.4. Consequences for erodibility assessment and erosion modeling
5. Conclusion
Acknowledgements
Chapitre 4 : Evaluation des facteurs explicatifs de variation de la stabilité structurale à pas de temps court
Short term dynamics of aggregate stability in the field : assessment of explanatory factors to improve erodibility prediction
Abstract
1. Introduction
2. Material and method
2.1. Sampling sites
2.2. Monitoring and sampling setup
2.3. Measurements
2.3.1. Aggregate stability
2.3.2. Variables linked to biological activity
2.3.3. Variables associated with climate
2.4. Statistical analysis
3. Results
3.1. Aggregate stability
3.1.1. Marcheville site
3.1.2. La Gouëthière site
3.2. Explanatory variables
3.2.1. Variables linked to biological activity
3.2.2. Variables linked to climate
3.3. Relationships between aggregate stability and explanatory variables
3.3.1. Relationships between aggregate stability and biological variables
3.3.2. Relationships between aggregate stability and climatic variables
3.4. Prediction of aggregate stability variations
4. Discussion
4.1. Factors linked to biological activity
4.2. Factors linked to climate
4.3. Prediction of aggregate stability and consequences for erosion predictions
5. Conclusion
Compléments à la partie 2
Synthèse et conclusion
Troisième partie – Caractérisation du réarrangement particulaire comme processus physico-chimique de variation de la stabilité structurale lié aux cycles d’humidité
Introduction
Chapitre 5 : Effet des cycles d’humectation-dessiccation sur la contrainte interne, le réarrangement particulaire et la stabilité structurale
Wetting and dryning cycle effects on internal stress, particle rearrangement and soil aggregate stability
1. Introduction
2. Material and method
2.1. Soil aggregate cylinder preparation
2.2. Wetting-drying cycle parameters
2.3. Measurements
2.3.1. Sensors
2.3.2. Aggregate stability
3. Results
3.1. Water tension
3.2. Vertical strain
3.3. Internal stress
3.4. Aggregate stability
4. Discussion
4.1. Influence of wetting and drying cycles on aggregate stability
4.2. Quantification of the variables measured by the sensors
4.3. Non reversible behaviour and particle rearrangement
4.4. Processes induced by wetting and drying cycles
5. Conclusion

GET THE COMPLETE PROJECT