Friends of Lake Merced 100 Santa Barbara Avenue Daly City, California 94014

 

December 27, 2000

Jim Hartley
Area Office Manager
CH2M-Hill
2485 Natomas Park Drive, Suite 600
Sacramento, CA 95833

Dear Mr. Hartley:

CH2M-Hill was retained by the San Francisco Public Utilities Commission (SF-PUC) in the mid-1990’s to support their effort to understand the relationship between Lake Merced and the Westside Basin Aquifer.  To that end you developed a model of the Westside Basin using a modeling routine called MicroFem.  The results of your study differed markedly from those of several prior studies, including those conducted by the United States Geologic Service (USGS) in the late 1980’s and Geo/Resource Consultants during the early ‘90s.

The most significant of these differences related to the water budget of Lake Merced.  The CH2M-Hill study concluded that groundwater flows play a relatively minor role in the lake water budget.  This conclusion has been seized upon by several major groundwater users in the Westside Basin (including golf courses located in the northern bounds of San Mateo County as well as the cities of Daly City and San Bruno) to support their contention that their pumping has little if any impact on Lake Merced.

The CH2M-Hill and Geo/Resources models both withstood the scrutiny of an expert peer review panel convened by the SF-PUC earlier this year.  However, that panel was provided little time and budget, and even less data, and as a result their assessment was necessarily general.  In an effort to further resolve questions regarding the fundamental lake budget we obtained additional information regarding input data, assumptions, and output of the CH2M-Hill model from Toni Pezzetti of your organization.  With the assistance of David Dawdy, retired USGS hydrologist, we have conducted an analysis of that information.  This analysis indicates that the CH2M-Hill model does not appear to be well calibrated, and that as a consequence several of the findings based upon this model are unreliable.  Specifically,  we have found what appear to be problems in the following areas:

1)  A comparison of measured and simulated water levels at calibration wells in the vicinity of Lake Merced reveals that the residuals are not random but are systematically biased.  The direction and location of these errors suggests that groundwater flow through Lake Merced has been underestimated and/or that the permeability of the aquitard separating the Colma and Merced formations has been underestimated.

2)  Discrepancies between simulated and measured lake level responses to the addition of surface water from other sources indicate that lakebed permeability has been underestimated.  As a result, when the addition of surface water to the lake is modeled, as occurred in the model year 1999, the resulting lake-level rise is maintained for several years.  However, measured lake levels in 1979 and 1994, years in which significant surface water was added to Lake Merced, show a recession to the prior lake level over a period of months.  Toni Pezzetti has observed that “no sustained benefit has been observed from periodic additions of surface water to Lake Merced.”  Actual observations and not simulated results lead to this conclusion.

3)  The extent and effect of the clay layer may be exaggerated in the model.   Borehole logs confirm that the aquitard is not laterally continuous between the Colma and Merced Formations, particularly near Lake Merced.  While your report states that “the cross sections indicate that the basin stratigraphy is highly variable,” that variability has not been reflected in the model.  Rather, the clay layer is modeled as a uniform, laterally extensive 50-foot-thick aquitard throughout the modeled area.

4) Assumptions used to estimate aerially-distributed recharge to the shallow aquifer from precipitation are inaccurate.  The model assumes that rainfall generates recharge every year.  In reality, deep percolation below the root zone in semiarid regions typically occurs only in normal to wet years, with no recharge in dry years.  It is more accurately simulated with a threshold function in an algorithm that tracks accumulated soil moisture storage on a daily or monthly basis throughout the simulation period.

5) There are similar issues with respect to deep percolation resulting from irrigation.  The model assumes that irrigation recharge is 50% of the recharge from precipitation.  Even with the assumption that irrigation recharge is proportional to precipitation recharge, this assumption reduces total recharge used in the model.  In addition, however, irrigation may be less in a wet year, reducing estimated recharge from this source.

6) The CH2M-Hill report downplays the importance of groundwater interaction with the lake, concluding that the relationship between Lake Merced and the supporting aquifer is “indirect”.  This conclusion is inconsistent with an internal CH2M-Hill memorandum from Nate Brown to Toni Pezzetti dated November 5, 1996 in which Mr. Brown observes that “Lake Merced is a surface water expression of the surrounding groundwater surface.  Thus, it is in direct hydraulic connection with the surrounding groundwater.”

Taken together these apparent problems raise substantial doubt as to the reliability of the CH2M-Hill model for estimating the effect of groundwater pumping or of other water management alternatives on the level of Lake Merced.  An accurate understanding of the interaction between Lake Merced and the surrounding aquifer system is fundamental to our efforts to manage lake levels and to protect beneficial uses.

Friends of Lake Merced will soon participate in an aggressive and well organized effort to enforce state regulations regarding the protection of Lake Merced.  In the course of that effort it may be necessary to challenge the results you have reported.  We believe a thorough review of the CH2M-Hill model is needed, if at all possible before that model is subjected to additional public scrutiny.

In the spirit of pursuing a cooperative effort we invite your assistance in addressing these important questions.  We look forward to working together with you to resolve these issues in a timely and professional manner.
 

Sincerely,
 

John Plummer
Friends of Lake Merced
 
 

c.c.: Michael Carlin
San Francisco Public Utilities Commission

David Dawdy
USGS – retired

Toni Pezzetti
CH2M-Hill

Dick Allen, Co-Chairman
Water Committee, Lake Merced Task Force

Tim Colen, Co-Chairman
Water Committee, Lake Merced Task Force