Mustafa, Golam2015-07-102015-07-101989http://hdl.handle.net/10919/54394This study presents a comprehensive physically based stochastic dynamic optimization model to assist planners in making decisions concerning mine soil depths and soil mixture ratios required to achieve successful revegetation of mined lands at different probability levels of success, subject to an uncertain weather regime. A perennial grass growth model was modified and validated for predicting vegetation growth in reclaimed mine soils. The plant growth model is based on continuous relationships between plant growth, air temperature, day length, leaf area, photoperiod and plant-soil-moisture stresses. A plant available soil moisture model was adopted to estimate daily soil moisture for mine soils. A general probability model was developed to estimate the probability of successful revegetation in a 5-year bond release period. The probability model considers five possible bond release criteria ir1 mine soil reclamation planning. A stochastic dynamic optimization model (SDOM) was developed to find the optimum combination of soil depth and soil mixture ratios that met the successful vegetation standard under non-irrigated conditions with weather as the only random element of the system. The SDOM was applied for Wise County, Virginia, and the model found that 2:1 sandstone/siltstone soil mixture required the minimum soil depth to achieve successful revegetation. These results were also supported by field data. The developed model allows the planners to better manage lands drastically disturbed by surface mining.viii, 169 leavesapplication/pdfen-USIn CopyrightLD5655.V856 1989.M877Reclamation of landRevegetationMine soilsStochastic processesDissertation