GMS:Automated Parameter Estimation: Difference between revisions
From XMS Wiki
Jump to navigationJump to search
No edit summary |
|||
| (6 intermediate revisions by 2 users not shown) | |||
| Line 3: | Line 3: | ||
An important part of any groundwater modeling exercise is the [[GMS:Model Calibration|model calibration]] process. In order for a groundwater model to be used in any type of predictive role, it must be demonstrated that the model can successfully simulate observed aquifer behavior. Calibration is a process wherein certain parameters of the model such as recharge and hydraulic conductivity are altered in a systematic fashion and the model is repeatedly run until the computed solution matches field-observed values within an acceptable level of accuracy. | An important part of any groundwater modeling exercise is the [[GMS:Model Calibration|model calibration]] process. In order for a groundwater model to be used in any type of predictive role, it must be demonstrated that the model can successfully simulate observed aquifer behavior. Calibration is a process wherein certain parameters of the model such as recharge and hydraulic conductivity are altered in a systematic fashion and the model is repeatedly run until the computed solution matches field-observed values within an acceptable level of accuracy. | ||
One of the tools provided in GMS for is automated parameter estimation. With automated parameter estimation, an external utility, sometimes called an "inverse model", is used to iteratively adjust a set of [[GMS:Parameters|parameters]] and repeatedly launch the model until the computed output matches field-observed values. Parameter estimation is used in conjunction with the [[GMS:Observations#Point Observations|point observations]] and the [[GMS:Observations#Flow Observations|flow observations]]. | One of the tools provided in GMS for is automated parameter estimation. With automated parameter estimation, an external utility, sometimes called an "inverse model", is used to iteratively adjust a set of [[GMS:Parameters|parameters]] and repeatedly launch the model until the computed output matches field-observed values. Parameter estimation is used in conjunction with the [[GMS:Observations#Point Observations|point observations]] and the [[GMS:Observations#Flow Observations|flow observations]] and is available for all MODFLOW versions that come with GMS except for [[GMS:MODFLOW-LGR|MODFLOW-LGR]]. | ||
Automated parameter estimation is supported in GMS for the [[GMS:MODFLOW|MODFLOW]] simulations using [[GMS:PEST|PEST]] a general purpose parameter estimation utility developed by John Doherty of Watermark Computing. | Automated parameter estimation is supported in GMS for the [[GMS:MODFLOW|MODFLOW]] simulations using [[GMS:PEST|PEST]] a general purpose parameter estimation utility developed by John Doherty of Watermark Computing. | ||
| Line 12: | Line 12: | ||
The basic steps involved in using an inverse model for parameter estimation are follows: | The basic steps involved in using an inverse model for parameter estimation are follows: | ||
; 1. Create a Working MODFLOW Model : The first step is to create a MODFLOW model and run a simulation. Before launching the inverse model, it's necessary to have a MODFLOW model that successfully converges and it's necessary to determine a good set of starting values for the parameters. Once there is a solution it is also a good idea to copy the computed heads from the solution to the starting heads array. This ensures that as the inverse model modifies the parameters and runs MODFLOW repeatedly, it is more likely that MODFLOW will quickly converge each time it is launched. | |||
The first step is to create a MODFLOW model and run a simulation. Before launching the inverse model, it's necessary to have a MODFLOW model that successfully converges and it's necessary to determine a good set of starting values for the parameters. Once there is a solution it is also a good idea to copy the computed heads from the solution to the starting heads array. This ensures that as the inverse model modifies the parameters and runs MODFLOW repeatedly, it is more likely that MODFLOW will quickly converge each time it is launched. | |||
; 2. Enter the Observations : Once there is a working MODFLOW model, enter the head and flux observations. [[GMS:Observations#Point Observations|Head observations]] are entered as points using an observation coverage in the Map module. [[GMS:Observations#Flow Observations|Flow observations]] are assigned directly to arcs and polygons in source/sink coverages. Each of the observations is assigned a [[GMS:Observations#Observation Weights|weight]] that is saved to the inverse input files. | |||
Once there is a working MODFLOW model, enter the head and flux observations. [[GMS:Observations#Point Observations|Head observations]] are entered as points using an observation coverage in the Map module. [[GMS:Observations#Flow Observations|Flow observations]] are assigned directly to arcs and polygons in source/sink coverages. Each of the observations is assigned a [[GMS:Observations#Observation Weights|weight]] that is saved to the inverse input files. | |||
; 3. Turn on the Inverse Model : Select an inverse model. Bring up the [[GMS:Global Options/Basic Package|''Global Options'']] dialog and select either the ''Parameter Estimation'' or ''Stochastic Inverse Model'' option depending on whether a stochastic simulation is being run. | |||
Select an inverse model. Bring up the [[GMS:Global Options/Basic Package|''Global Options'']] dialog and select either the Parameter Estimation or Stochastic Inverse Model | |||
; 4. Parameterize the Model : The next step is to parameterize the model. See the [[GMS:Parameters#Parameterizing_the_model|Parameters]] page for more details. | |||
The next step is to parameterize the model. See the [[GMS:Parameters#Parameterizing_the_model|Parameters]] page for more details. | |||
; 5. Create a Parameter List : The next step is to create the parameter list. See the [[GMS:Parameters#Create_a_parameter_list|Parameters]] page for details. | |||
The next step is to create the parameter list. See the [[GMS:Parameters#Create_a_parameter_list|Parameters]] page for details. | |||
; 6. Set Parameter Estimation Options : Once the parameter list is set up, the user may wish to edit the general [[GMS:PEST Dialog|''Parameter Estimation'']] options. These options include the output control and convergence criteria. | |||
Once the parameter list is set up, the user may wish to edit the general [[GMS:PEST Dialog|''Parameter Estimation'']] options. These options include the output control and convergence criteria. | |||
; 7. Edit the Group Weight Multipliers : The [[GMS:Observations#Group Weight Multipliers|group weight multipliers]] can be edited to adjust the relative weight of the head and flux observations. | |||
The [[GMS:Observations#Group Weight Multipliers|group weight multipliers]] can be edited to adjust the relative weight of the head and flux observations. | |||
; 8. Edit the PEST ASP Package : Edit the MODFLOW [[GMS:PEST ASP Package|PEST ASP Package]] if necessary in order to ensure a stable solution. | |||
Edit the MODFLOW [[GMS:PEST ASP Package|PEST ASP Package]] if necessary in order to ensure a stable solution. | |||
; 9. Save and Run MODFLOW Model : Once all of the inverse model options have been set, the next step is to save the MODFLOW model using the '''Save''' or '''Save As''' command in the ''File'' menu. Next, run MODFLOW and the inverse model will run with MODFLOW. The inverse model will then be launched in a separate window or the model wrapper in which the user should see information relating to the MODFLOW runs and the status of the objective function. Depending on the problem, the inverse model may take anywhere from several minutes to several hours (or days) to run to completion. When the inverse process is completed successfully, GMS automatically launches a MODFLOW forward run with the optimal values computed by the inverse model. Thus, the solution will reflect the optimal values computed by the inverse model. | |||
Once all of the inverse model options have been set, the next step is to save the MODFLOW model using the '''Save''' or '''Save As''' command in the ''File'' menu. Next, run MODFLOW and the inverse model will run with MODFLOW. The inverse model will then be launched in a separate window or the model wrapper in which the user should see information relating to the MODFLOW runs and the status of the objective function. Depending on the problem, the inverse model may take anywhere from several minutes to several hours (or days) to run to completion. When the inverse process is completed successfully, GMS automatically launches a MODFLOW forward run with the optimal values computed by the inverse model. Thus, the solution will reflect the optimal values computed by the inverse model. | |||
; 10. Viewing the Optimal Values : When the inverse model is finished, it writes out a text file containing the set of parameter values corresponding to the minimum calibration error. These values can be viewed with the '''Import Optimal Values''' button. This copies the optimal parameter values to the ''Starting Value'' field in the ''Parameter List''. | |||
When the inverse model is finished, it writes out a text file containing the set of parameter values corresponding to the minimum calibration error. These values can be viewed with the '''Import Optimal Values''' button. This copies the optimal parameter values to the ''Starting Value'' field in the ''Parameter List''. | |||
==Sensitivity Analysis== | ==Sensitivity Analysis== | ||
| Line 47: | Line 37: | ||
For a more detailed description of parameter sensitivity see section 5.3.2 of the PEST manual. | For a more detailed description of parameter sensitivity see section 5.3.2 of the PEST manual. | ||
== | ==PEST Dialog== | ||
Options affecting parameter estimation can be changed via the [[GMS:PEST Dialog|'' | Options affecting parameter estimation can be changed via the [[GMS:PEST Dialog|''PEST'' dialog]]. | ||
==Parallel PEST== | ==Parallel PEST== | ||
The PEST model can be run with parallel processing across one or multiple machines with the | The PEST model can be run with parallel processing across one or multiple machines with the parallel PEST utility. Running across multiple machines requires setup outside of GMS with PSLAVE. More information about parallel PEST can be found in the [http://www.pesthomepage.org/getfiles.php?file=newpestman1.pdf PEST Manual Part I] starting on page 222 from the [http://www.pesthomepage.org/Downloads.php PEST Downloads page]. | ||
==See also== | ==See also== | ||
| Line 63: | Line 53: | ||
[[Category:PEST]] | [[Category:PEST]] | ||
[[Category:Parameters]] | [[Category:Parameters]] | ||
[[Category:External Links]] | |||
