Observation sensitivities are used to calculate a number of statistics that can be used to diagnose inadequate data, to identify parameters that probably cannot be estimated by regression using the available observations, and to evaluate the utility of proposed new data. These are called observation sensitivities. If the Observation Process is active, it uses the grid sensitivities to calculate sensitivities for the simulated values associated with the observations. The Sensitivity Process calculates the sensitivity of hydraulic heads throughout the model with respect to specified parameters using the accurate sensitivity-equation method. In addition, a number of files are produced that can be used to compare the values graphically. A variety of statistics is calculated to quantify this comparison, including a weighted least-squares objective function. The Observation Process generates model-calculated values for comparison with measured, or observed, quantities. This PDF technical document describes the Observation, Sensitivity, and Parameter-Estimation Processes of the groundwater modeling computer program MODFLOW 2000. MODFLOW 2000 - User Guide to Observation, Sensitivity, Parameter-Estimation Processes, and Three Post-Processing Programs wise use of statistics generated using calculated sensitivities and the match between observed and simulated values, and associated graphical analysesįourteen guidelines presented in this work suggest ways of constructing and calibrating models of complex systems such that the resulting model is as accurate and useful as possible.defining a tractable inverse problem using simplifications appropriate to the system under investigation and.Inverse modeling in many fields is plagued by problems of instability and nonuniqueness, and obtaining useful results depends on: Minimization is accomplished using a modified Gauss-Newton method, and prior or direct information on estimated parameters can be included in the regression. UCODE and MODFLOWP perform inverse modeling by calculating parameter values that minimize a weighted least-squares objective function using nonlinear regression. The inverse modeling and statistical methods discussed are broadly applicable, but are presented as implemented in the computer programs UCODE and MODFLOWP. This PDF technical document describes methods and guidelines for model calibration using inverse modeling.
0 kommentar(er)
