Page 102 - California Stormwater Workshop Handouts

Industrial General Permit Fact Sheet

In 2006, The U.S. Court of Appeals for the Sixth Circuit held that the CWA does not
require U.S. EPA to set numeric limits where such limits are infeasible. (Citizens
Coal Council v. United States Environmental Protection Agency, 447 F.3d 879, 895-
96 (6th Cir. 2006)). The Citizens Coal court cited to the statement in Waterkeeper
Alliance, Inc. v. EPA, 399 F.3d 486, 502 (2d Cir. 2005) that “site-specific BMPs are
effluent limitations under the CWA” in concluding that “the EPA's inclusion of
numeric and non-numeric limitations in the guideline for the coal remining
subcategory was a reasonable exercise of its authority under the CWA." (447 F.3d
at 896.) Additionally, the Citizen’s Coal court cited to Natural Res. Def. Council, Inc.
v. EPA, 673 F.2d 400, 403 (D.C.Cir.1982) noting that “section 502(11) [of the CWA]
defines ‘effluent limitation’ as ‘any restriction’ on the amounts of pollutants
discharged, not just a numerical restriction.” NPDES permit writers have substantial
discretion to impose non-quantitative permit requirements pursuant to section
402(a)(1)), especially when the use of numeric limits is infeasible. (NRDC v. EPA,
822 F.2d 104, 122-24 (D.C. Cir. 1987); 40 C.F.R. 122.44(k)(3).)

4. Decision to Include Non-Numeric Technology-Based Effluent Limits in This General
Permit

It is infeasible for the State Water Board to develop numeric effluent limitations using
the best professional judgment approach due to lack of sufficient information.
Previous versions of this General Permit required Dischargers to sample their
industrial storm water discharges and report the results to the Regional Water
Boards. Dischargers were not required to submit this data online into a statewide
database; as a result, much of this data is not available for analysis. Moreover,
much of the data that are available for analysis are not of sufficient quality to make
conclusions or perform basic statistical tests.

The Blue Ribbon Panel of Experts, State Water Board staff, and many stakeholders
evaluated the available storm water data set and concluded that the information
provides limited value due to the limited pool of industrial facilities submitting data,
poor overall data quality, and extreme variance within the dataset, as described
below.

The poor quality of the existing data set is attributable a number of factors. For
example, the previous permits have required Dischargers to sample during the first
hour of discharge from two storm events a year. This sampling schedule was
designed to catch what was considered to represent the higher end of storm water
discharge concentrations for most parameters. The results from this type of
sampling were thought to be an indicator of whether or not additional BMPs would
be necessary. The sampling schedule was not designed, however, to estimate
pollutant discharge loading, or to characterize the impact of the discharge on the
receiving water. Doing so would normally require the use of more advanced
sampling protocols such as flow meters, continuous automatic sampling devices,
certified/trained sampling personnel, and other facility-specific considerations.

Furthermore, there is currently no data which details the relationship between the
BMPs implemented at each facility and the facility’s sampling results. The SWPPPs
required by the previous permits were not submitted to the Water Boards, but were

Order 2014-0057-DWQ 18