Abstract: Artificial neural networks (ANNs) are used to estimate vertical ground surface movement when soils expand and contract due to changes in soil moisture content caused by changing climate conditions. Several counterpropagation ANN test cases were investigated to map climate data (i.e. temperature and rainfall) to vertical ground surface movement at field sites in Texas and Australia. Three of the four ANN test cases use a historical time series of climate data to forecast ground surface elevation relative to a specified datum. The fourth ANN test case predicts the rate of ground surface movement, and requires post-processing of the predicted rates to calculate ground surface elevation relative to a specified datum. The counterpropagation network has demonstrated a successful mapping of temperature and rainfall data to vertical ground surface movement at a field site when it is trained with a subset of data from the same field site (test cases 1 and 2). The results of training an ANN on one field site and testing it on another field site (test cases 3 and 4) demonstrate the ability of the ANN to capture trends in vertical ground surface movement. When compared with the predictions from a physics-based method (shrink test-water content method) that requires measurements/estimates of changes in soil water content, the ANN-based predictions (based on climatic changes) captured the trends in the field measurements of shrinking–swelling soil surface movements equally well. These findings are promising and merit further investigation with data from additional field sites.
Abstract: Artificial Neural Networks (ANNs) are used to predict vertical surface movement when soils expand and contract with changes in soil moisture caused by climatic conditions. Temperature and rainfall data, soil property data, and soil moisture measurements are used for training ANNs to simulate the movement of spread footings at a field site in Arlington, Texas. A research team from Texas A&M University surveyed the footing movement monthly over a two-year period. The performance of the ANNs is evaluated by comparing the predictions to the observed movements. Data for temperature and rainfall are available for each month of the two-year field study, but soil moisture data (on which soil shrinking/swelling is predominantly dependent) is only available for seven of those months. Therefore, two ANNs were used in series. The first ANN estimates soil moisture for the months when data are not available, while the second ANN uses both measured and estimated soil moisture as training input to estimate vertical movement.
Abstract: A hierarchical system of simple, geostatistical-based, artificial neural networks (ANNs) have been developed to enhance existing geographic information system (GIS)-based watershed management tools for diagnosing geomorphic instability at a variety of sub-basin and watershed scales. Two ANNs originally developed for the classification of reach-scale vulnerability and geomorphic condition have been tested (in concert with best judgment by experts) using existing data for two Vermont watersheds. These ANNs will support future development of modules to enhance land use management at the watershed scale to better predict geomorphic instability and sediment transport in response to natural and anthropogenic stresses.
Abstract: An artificial neural network (ANN) is used to predict structural movement at a specific site using available soil data at the site and in the vicinity. ANNs are data-driven computational tools that map data inputs, such as moisture content and shrink/swell properties of soils, to desired outputs, such as soil displacement. The performance of the ANN is evaluated by comparing its predictions to observed structural movement. Although the back-propagation algorithm is the most commonly used training technique for ANNs, it is not always the most efficient. Other ANN architechtures, based on generalized regression and counter-propagation algorithms are more efficient for some problems, and are tested for this problem.