Synthetic Groundwater Model Calibration Exercise

(Advanced – Multi-parameter estimation/calibration; pre- and-post pumping analysis)

Figure 1: Site of interest. Length in x- (horizontal) direction: 3000m.

Background:

Planning to install a new 250 GPM well at their main facility, a small manufacturer hires you to determine whether their proposed well (see above) will cause any problems with the stream and an existing well and whether the new well or the existing well would likely become contaminated by a nearby TCE plume as a result of the operation of the new well.  The new well will be operated continuously.

Geologic data indicate a water table aquifer with bottom elevation of –100ft in this area.  The aquifer abruptly pinches out at west and southern boundaries. Ground surface elevation in the area is approximately 3m. A 1^st-order stream (average depth=1m) flowing through the area empties into the deep lake (water surface elevation 0.00m) in the NE corner of the map.  The stream is predominantly fed by groundwater, meaning the streamflow is essentially baseflow.

Although there has been no study of actual recharge rates to this aquifer, long term average recharge from precipitation in this part of the state varies between 10 and 18 inches/yr depending on the location of the aquifer. 

Only one well in the vicinity currently pumps (200 GPM) from this aquifer. This well has been operating continuously for the past 15 years.

This past August, in anticipation of the need to understand this aquifer better, 45 piezometers were installed to determine the water table variation throughout the area.  In addition, stream water levels and stream discharges were measured.  Average stream discharge measured at a cross section approaching the lake was 1486.65 m³/d (272.73 GPM). The streambed leakance (hydraulic conductivity per unit thickness of the streambed) is unknown and must be calibrated. Annual average water table elevations and stream water levels are provided below (see Appendix). 

The party responsible for the TCE plume has only recently been notified that they must remediate this plume.  The State Department of Environmental Quality expects that it will be several years before this happens since it appears the State will have to go to court to force compliance. 

Objective:

Assess the likely impact of the new well based on the data provided and prepare a 1-2-page memo to me that summarizes your approach and findings. You should also provide final model inputs and calibration parameters and detailed model results in an appendix to allow me to follow your analysis and the basis for your conclusions. 

Specific steps required for this analysis:

• Model Set-up
• Calibration
• Application

MAGNET MODELING Hints/Suggestions:

• Setup the domain to have a length of 3000m in the x- (horizontal) direction and 2230m in the y- (vertical) direction. You may directly load the DomainTextFile (DM from a Txt File) available with this problem on the MAGNET Curriculum Network. Or go to Synthetic Case mode (Other Tools > Utilities > Go to Synthetic Case Area) and adjust the domain to the appropriate size.
• The Site Map image (Figure 1) and Monitoring Wells map (Figure 2) can be overlaid to the map display. Use the image files available with the problem on the MAGNET Curriculum Network Overlay and the myImage tool (under ‘Utilities’). Choose ‘Use Domain Extent’ to assign the image extents after uploading the appropriate file.
• Use a large grid size (NX=100) to resolve the drawdown near the proposed pumping well.
• Calibrate to conditions before the proposed well is added to model
  • Calibration parameters: hydraulic conductivity, recharge, and streambed leakance
  • Calibration targets: observed head data and observed streamflow
• Calibrate the model to observed head data by using the Calibration tool and the stream flow data by performing a water budget analysis (Analysis Tools > Analysis > Display Charts).
• Using the Calibration tool (Analysis Tools > Calibration):
  • First, overlay the Monitoring Wells image to the Map Display
  • Select the Calibration tool; answer 'Cancel' to the first prompt that appears to compare simulated and observed head across the entire model domain.
  • Select 'Table' from the drop-down menu under Observed Data Sources. This opens the Calibration Data Input interface with a blank data table. 
  • Click the 'Click to Add Data' button to use the cursor to place observation points. When you click on the map, a marker is added, and a row entry is added to the table.
  • Update the ‘V’ (Value) entry (2^nd to last column in each row) with observed water level data. It is also helpful to re-label each well to match the numbering scheme used below (1^st column – WellID). 
  • Use the 'Click to Add Data' button to add another data point. It may be difficult to “see” each well on the provided site map, so do your best to add as many wells as you can. An incomplete but representative observation data set is sufficient for calibration.
     
• Apply particle tracking from the contaminated site under remediation after the proposed well is added to the model

Appendix: Field Data

Observed Water Level Data

Figure 2: Monitoring well network.

NOTE: Data for all 45 wells are provided below, but it may be difficult to “see” each well on the provided site map. Do your best to add as many wells as you can. An incomplete but representative observation data set is sufficient for calibration.

Monitoring Well Observations 

Stream Stage Data

Figure 3: Stream stage measurement locations

Measured stream stages (elevations)