The Kleinfelder Report suffers
two main problems (1) it deviates from the proposed scope of services,
and (2) it deviates
from basic hydrogeologic concepts by dwelling on a statistical analysis
of “depth to water trends”.
The following are deviations from the scope of services:
1. In data collection and assessment, Kleinfelder
failed to collect data for two DWR monitoring wells in the Bennett
Valley Study Area and 14 private water company wells in or very near
Bennett Valley and Mark West Springs Study Areas. These data are
easily accessible on the internet and, therefore, Kleinfelder failed
to properly assess ease of data collection. (See Task Two and Task
Six, item 3)
2. In relating geology and hydrogeology to water
quality, Kleinfelder failed to utilize readily available water quality
data from the 14 private water company wells mentioned above. (See
Task Five, Technical Component; Task 6, item 5)
3. Kleinfelder did not provide any useful information
on groundwater recharge. (See Task Five, Technical Component; Task
6, item 4)
4. Kleinfelder failed to assess how different entities
affect each other’s water availability. Specifically, Kleinfelder
did not distinguish between residential usage by single-property
wells and usage by private water company wells for subdivisions.
Kleinfelder did not assess how usage for new developments, particularly
residential subdivisions and golf courses, will affect water availability
for prior users, primarily single residential units and agriculture.
(See Task Five, Water Use Component)
5. Kleinfelder did not assess or address environmental
constraints, such as maintaining flows in Salmon Creek and Mark West
Creek, both recognized as salmonid habitat. (See Task Five, Future
Demand Component)
6. Kleinfelder failed to assess changing land use
in Mark West and Bennett Valley, particularly in the last decade,
favoring residential subdivisions and golf courses. Kleinfelder examined
no hydrology data from after 1992. (See Task Five, Land Use Component)
Kleinfelder deviated from basic hydrogeologic concepts by dwelling on a statistical
analysis of “depths to water trend” for its primary technical
contribution. This approach is fraught with errors related to:
- Topography (a major consideration for all three
Study Areas),
- Development trends toward ridgetops,
- Changes in drilling technology facilitating
deeper drilling,
- Trends from drilling shallow wells serving single
properties to drilling deep high-capacity, public water supply
wells serving subdivisions.
- Changes in land use, such as converting open
space to golf courses (which Kleinfelder showed to be the largest
groundwater users by far).
Kleinfelder should have focused on establishing
change in water level elevations over time. Water level elevations,
not depths, establish the volume of groundwater stored, which is
crucial in the Joy Road area where the Wilson Grove Formation is
perched on top of the Franciscan Formation. Water level elevations
dictate the rate and direction of flow. If deep large capacity wells
for golf courses and subdivisions drop water level elevations too
much, shallow wells will go dry and summer creek flows will cease. |
|
When compared to the revised Scope of Work, the Kleinfelder
pilot project report contains virtually none of the promised results.
In particular, the data have not been used,
as proposed, to calculate water budgets for each area, to show
the "current groundwater
elevations, directions of water movement, and quality, or " ...
an analysis of the "rate of increased groundwater extraction" versus
water availability. A summary memorandum to discuss "hydrogeology
relating to water quality, quantity, recharge, and consumption rates,
[and] environmental regulations," is also missing.
According to the report, Kleinfelder "prepared GIS maps and databases
to represent and analyze the collected data." Sadly, the
power of a GIS database to overlay and compare multiple data
set at the
same scale is not demonstrated in this report.
Especially disappointing is the large number of factual and data
presentation mistakes.
The improper data analysis is a more critical flaw.
Variable scaling in single graphs, and (or) mistakes in labeling
the graphs raise other questions about Kleinfelder's competence
and technical
capacity. It is hard to believe that a computer program could
plot such inconsistently variable graph scales (or were they
not computer-plotted?).
Reviews by Kleinfelder's Senior Project Manager, a California
Registered Geologist, should insure that the text and data
analyses exhibit
a high level of expertise and technical competence, but expertise
and
competence are not very evident to the trained reader.
The Plates contain information missing from figures, but
are not all plotted at the same scale, or with the same projection
so that
they
cannot even be overlaid on a light table, to make comparisons
between natural and cultural features. Only one diagram (Fig.
24) shows
the congruence between settlement patterns and geology.
As required by the revised Scope of Work, the most important
product of this pilot study was to be a computer database,
compatible with
the County's Geographic Information System. Unfortunately,
the data reduction and graphical representations of yearly
precipitation,
well
water-level, and well depth data, in Figures 1 through 10
are multiply flawed. A major flaw is the supposed lack of
well
location data,
which prevents conversion of well-depth data to elevations,
which is the
standard for professional groundwater level analysis. Lacking
locations, the well water and depth data cannot be examined
in relation to
the geologic units in which they are drilled.
Although the lack of most location data would render the
study nearly valueless, this excuse is insupportable. The
data were
obtained from
state files, which assign a standardized well identification
number to each well. These standardize IDs consist of location
information
detailed enough to place any well within a 40-acre area.
Ignoring the location data in their hands Kleinfelder, crudely
averaged well and water depth values for each year represented.
Wells drilled
into aquifer rocks are lumped with those in aquitard rocks,
and the depths of wells on ridges are also combined with
those in
valleys and
on hillslopes, creating graphs that could be said to represent
apples, oranges, and grapefruit.
As may be expected from graphs that average unlike data,
these contain obvious distortions. Combinations of unrelated
data
types are the
probable explanation for incoherencies between low well water
levels and high
prior-year precipitation in the Figure 1 and 2 graphs (explained
on p. 35 as due to "other variables ... also at play.").
Averaging such disparate data likely obscures actual water
level changes in the
study areas.
Another problem is the absence of very recent water or well
data in this study the most recent are at least 5 (or 10)
years old
mistakes
in graph scales, or labels, and missing dates in the data
tabulations, make it difficult to impossible to determine
where the problems
lie. Only the tabulated Joy Ridge area data include dates.
The worst data-handling error in this report are plots of
well water depths from an unidentified surface, ignoring
topography.
The lack
of well locations apparently made it impossible to calculate
and plot any but a few water levels as elevations. Thus,
there are
no plots
of the groundwater levels as elevation ("piezometric
surfaces")
for any of the study areas. This omission seems inexplicable
and does not accord with "generally accepted standards
of care" that
I am aware of. (continued . . . download the entre document) |
