An expert system for predicting the infiltration characteristics

Infiltration plays a fundamental role in streamflow, groundwater recharge, subsurface flow, and surface and subsurface water quality and quantity. This study includes a comparative analysis of the two machine learning techniques; M5P model tree (M5P) and Gene Expression Programming (GEP) in predictions of the infiltration characteristics. The models were trained and tested using the 7 combination (CMB1 – CMB7) of input parameters; moisture content (m), bulk density of soil (D), percentage of the silt (SI), sand (SA) & Clay (C), and time (t), with output parameters; cumulative infiltration (CI) and infiltration rate (IR). Results suggested that GEP has an edge over M5P to predict the IR and CI with R, RMSE & MAE values 0.9343, 15.9667 mm/hr & 8.7676 mm/hr, and 0.9586, 9.2522 mm and 7.7865 mm for IR and CI, respectively with CMB1. Although the M5P model also gave good results with R, RMSE & MAE values 0.9192, 14.1821 mm/hr, & 19.2497 mm/hr, and 0.8987, 11.2144 mm & 18.4328 mm for IR and CI, respectively, but lower than GEP. Furthermore, single-factor ANOVA and uncertainty analysis were used to show the significance of the predicted results and to find the most efficient soft computing techniques respectively.

Development and Calibration of Automated Multiple-Ring Infiltrometer

An automatic triple-ring infiltrometer was developed using a set of pre-set sensors and transducer (AP 403, AP 404, AP 405 and AP 406, RAP001 and RAP002). The aluminum probe sensors were graduated and arranged in series to monitor the rate at which water is infiltrating into the soil layer. The working principle of automatic triple-ring infiltrometer was developed using six probes with depth calibration of 1.0mm, 26.7 mm, 12.4 mm, and 12.7 mm, respectively. The result obtained showed strong agreement with a coefficient of determination R2= 0.963, indicating positive proportionality between cumulative infiltration and time taken for the water to infiltrate at different depths. The instrument has a measuring accuracy of ± 0.3mm infiltration depth. The device works effectively under biochar amended soil and other soil formations with high precision. Accurate infiltration data generated by the instrument would be applied to estimate the depth of water available to plant and predict possible agricultural drought.

Development and Calibration of Automated Multiple-Ring Infiltrometer

An automatic triple-ring infiltrometer was developed using a set of pre-set sensors and transducer (AP 403, AP 404, AP 405 and AP 406, RAP001 and RAP002). The aluminum probe sensors were graduated and arranged in series to monitor the rate at which water is infiltrating into the soil layer. The working principle of automatic triple-ring infiltrometer was developed using six probes with depth calibration of 1.0mm, 26.7 mm, 12.4 mm, and 12.7 mm, respectively. The result obtained showed strong agreement with a coefficient of determination R2= 0.963, indicating positive proportionality between cumulative infiltration and time taken for the water to infiltrate at different depths. The instrument has a measuring accuracy of ± 0.3mm infiltration depth. The device works effectively under biochar amended soil and other soil formations with high precision. Accurate infiltration data generated by the instrument would be applied to estimate the depth of water available to plant and predict possible agricultural drought.

Effects of Biochar on Infiltration and Evaporation of Soil Water with Sand Mulching

Water shortages and soil salinization in arid regions of northwestern China have become an important factor limiting agricultural production and affecting the living conditions of farmers. Using existing water resources and salinized land scientifically and efficiently is important for improving agricultural production. We studied the characteristics of the infiltration and evaporation of soil water under the application of different amounts of biochar in indoor soil-column simulations with sand mulching to determine the effect of biochar on the infiltration and evaporation of soil water. We tested five treatments: sand mulching with no biochar (CK) and sand mulching with 0.5, 1.5, 2.5 and 4.5% biochar. Cumulative infiltration and the rate of migration of the wetting front were higher with 0.5 and 1.5% biochar and lower with 2.5 and 4.5% biochar than CK, respectively. The distance and duration of migration of the wetting front followed the power function F = a·tb, a relationship consistent with the Kostiakov model of infiltration. The cumulative evaporation of soil water was higher with 0.5 and 1.5% biochar and lower with 2.5 and 4.5% biochar than CK. The variation of cumulative evaporation was represented well by both the Black and Rose models of evaporation. The results of this study have important implications for the application of biochar in arid regions of northwestern China for improving the hydrological characteristics of soil in agricultural practices.

Responses of Soil Infiltration to Water Retention Characteristics, Initial Conditions, and Boundary Conditions

(1) Background: Simulation of soil water infiltration process and analysis of its influencing factors are important for water resources management. (2) Methods: In this study, the relative contributions of the soil water retention characteristics (SWRC) estimation, initial water content, and constant pressure head at upper boundary to the cumulative infiltration under various soil conditions were quantified based on the 1-D Richards’ equation and 900 scenarios. Scenario simulations were performed for two SWRC estimation methods (Jensen method and Rosetta); three different initial water contents (0.15, 0.20, and 0.25 cm3/cm3); five different constant pressure heads (0.5, 1, 2, 4, and 8 cm); and thirty soil samples with varying texture and bulk density. (3) Results: Rosetta representing the drying branch of the SWRC yielded higher simulated cumulative infiltration compared with the Jensen method representing the wetting branch of the SWRC. However, the Jensen method–predicted cumulative infiltration fluxes matched well with the measured values with a low RMSE of 0.80 cm. (4) Conclusions: The relative contribution of the SWRC estimation method to cumulative infiltration (19.1–72.2%) was compared to that of constant pressure head (14.0–65.5%), and generally greater than that of initial water content (2.2–29.9%). Findings of this study have practical significance for investigating the transport of water, nutrients, and contaminants in the unsaturated zone.

Coupling large and small ring infiltration experiments for investigating preferential flow

<p>Preferential flow is more the rule than the exception. Water infiltration is often led by preferential flow due to macropores, specific soil structures (e.g., aggregates, macropore networks), or lithological heterogeneity (occurrence of materials with contrasting hydraulic properties). Water infiltration in soils prone to preferential flow strongly depends on soil features below the soil surface, but also the initiation of water infiltration at the surface. When the macropore networks are not dense, with only a few macropores intercepting the soil surface, water infiltration experiments with ring size in the order of 10-15 cm diameter may overlook sampling macropore networks during some infiltration runs, minimizing the effect of macropore flow on the bulk water infiltration at the plot scale.</p><p>In this study, we investigated the effect of ring size on water infiltration into soils prone to preferential flow. We used two ring sizes: small (15 cm in diameter) and large (50 cm in diameter). By doing so, we hypothesized that the large rings, sampling a more representative soil volume, will maximize the probability to intercept and activate a macropore network. In contrast, the small rings may activate the macropore network only occasionally, with other infiltration runs mainly sampling the soil matrix. Thus, the small rings are expected to provide more variable results. On the other hand, the large rings are expected to provide more homogeneous results in line with the soil's bulk infiltration capability, including all pore networks at the plot scale.</p><p>Three different sites were sampled with varying types of preferential flow (macropore-induced versus lithological heterogeneity induced). The experimental plan included inserting large rings at several places in the experimental sites with a dozen small rings nearby to sample the same soil. All the rings were submitted to a similar positive constant water pressure head at the soil surface. The cumulative infiltrations were then monitored and treated with BEST algorithms to get the efficient hydraulic parameters. Note that the cumulative infiltration could not be compared directly since lateral water fluxes varied in extent and geometry between the different ring sizes. The impacts of the ring size on the magnitude of cumulative infiltration and related estimated hydraulic parameters were discussed. Our results demonstrated the impact of ring size but also the dependency of such effect on the site and the type of flow.</p><p>Our results contribute to understanding preferential flow in heterogeneous soils and the complexity of its measure using regular water infiltration devices and protocols.</p>

Predicting the infiltration characteristics for semi-arid regions using regression trees

The study of infiltration process is considered as essential and necessary for all hydrology studies. Therefore, accurate predictions of infiltration characteristics are required to understand the behavior of subsurface flow of water through the soil surface. The aim of the current study is to simulate and improve the prediction accuracy of infiltration rate and cumulative infiltration of soil using regression tree methods. Experimental data recorded with a double ring infiltrometer for 17 different sites are used in this study. Three regression tree methods: Random tree, Random forest (RF) and M5 tree are employed to modelling the infiltration characteristics using the basic soil characteristics. The performance of the modelling approaches is compared in predicting the infiltration rate as well as cumulative infiltration, obtained results suggest that performance of RF model is better than other applied models with coefficient of determination (R2) = 0.97 & 0.97, root mean square error (RMSE) = 8.10 & 6.96 and mean absolute error (MAE) = 5.74 & 4.44 for infiltration rate and cumulative infiltration respectively. RF model is used to represent the infiltration characteristics of the study area. Moreover, parametric sensitivity is adopted to study the significance of each input parameter in estimating the infiltration process. Results suggest that time (t) is the most influencing parameter in predicting the infiltration process using this data set.