Robert P. Schreiber is an MIT-trained engineer and specialist in wastewater flow. This is a summary of his analysis of Leyland’s latest proposal to the DEP.

Memorandum

To: SGOD Membership

From: Robert P. Schreiber, P.E., DEE

Date: September 30, 2005

Subject: Madison Landing – BLC Report – Review Comments

Per your request, I have reviewed the letter-report of August 18, 2005 authored by BL Companies (BLC), and offer the following comments. The first section below presents a summary of my review comments……

Section 1 – Summary of Review Comments

1. All of my prior technical review comments still apply. BLC has not provided any new data to prove their conceptual model and theories. In addition, this latest round of modeling does not include any significant improvements over prior modeling, and in fact appears to have produced more questionable results. I believe that a significantly more conservative approach is needed because of the uncertainties involved and the indications that much lower permeabilities prevail at the site and the hydraulic boundaries.

2. The groundwater flow modeling is still based on a conceptual model that is not fully backed up by data. BLC could have taken this opportunity to obtain the critical data that I outlined in a prior submittal. Instead, BLC has continued to perform simulations based on the same unproven theories and assumptions.

3. The use of an average K value of 200+ or 500+ feet per day for the entire thickness of the top layer in the model is probably a significant overestimation of the layer’s transmissivity – based on log after log showing the predominance of much finer sediments and stratification, as well as the absence of any logs whatsoever between the site and Long Island Sound.

4. Because the model does not simulate the strong gradient at Bradley Creek – right next to the proposed infiltration system -- the model can easily include a higher permeability (and rainfall-recharge rate) than is actually the case. In turn, the model will predict a significantly smaller amount of water table mounding than in reality.

 

5. Similarly, the model simulates water table elevations in the salt marsh that are in some cases 2 to 4 feet above the ground surface of the marshes. The model has in effect produced a ponded marsh area, with up to 4 feet of standing water on average – just as it simulated up to 5 feet of standing water on average in Bradley Creek. This is a clear misrepresentation and another indication of why the model cannot be used to predict water table mounding or flow patterns accurately.

6. Using industry-standard guidance for groundwater flow model calibration, the model appears to exhibit fair to poor calibration statistics.

7. In addition to the overall measures of calibration that indicate inadequacy, the predictive simulations produced results that point to a systematic flaw. Specifically, the lower permeability model (Scenario 1) produced predicted water table mounding that is only slightly greater than the higher permeability model’s prediction. This does not make sense.

8. The specific capacity values from the site’s pumping tests were very low, indicating much lower transmissivity than simulated in BLC’s model. BLC should have calibrated their latest model to the pumping test results. Based on their past efforts in this regard, the approach used was inappropriate, leading to significant overestimation of the prevailing transmissivity at the site.

9. BLC’s model downplays the possibility of flow toward the west because of their assumption that a fixed water table divide exists. By not taking the more reasonable and accurate approach that the water table does move, and would likely be defined by the effluent loading, BLC’s model underestimates the possibility and magnitude of flow toward the closest property boundary points – Young’s Village and Bradley Creek.

10. For performing a complete technical review, significantly more data and information from the modeling needs to be obtained – including the model input files and results, with detailed information on the discharges to various features.

11. BLC downplays the potential concerns regarding additional freshwater and nutrient discharges to the salt marshes, by indicating that a fragmites [sic] control program will suffice. However, this is much too casual of an approach to a situation that may not be nearly so simple – especially in light of the likely underestimation of the percentage increase in freshwater discharge and nutrient delivery to the nearby salt marshes.