Air Insulated Substation or Gas Insulated Substation
SDG&E proposed to build a conventional Air-Insulated Substation (AIS), but in its PEA, SDG&E also proposed an alternative of a Gas-Insulated Substation (GIS). The Final EIR found the GIS alternative to be environmentally preferred. However, SDG&E strongly prefers to build an AIS facility. Accordingly, we need to evaluate whether we should authorize construction of a GIS or an AIS at the Silvergate site. In performing that evaluation, we will consider the environmental impact, cost, and timing of each.
According to the Final EIR, the environmental benefits of the GIS design stem from its smaller footprint (up to approximately 4.3 acres), as compared to an AIS design (approximately five acres). Because of its smaller size, the GIS design could be built without the acquisition and demolition of an existing building owned by Propulsion Controls Engineering (PCE). The Final EIR finds, based on SDG&E's PEA, that avoiding acquisition and demolition of the PCE building results in reduced environmental impacts in the areas of Land Use; Air Quality, Hydrology and Water Quality, and Hazardous Materials; and Visual Impacts. (Final EIR, pp. 3C1-19 to 3C1-20.)
The reduced impact in the area of Land Use comes from avoiding the need to relocate the existing business. The reduced impacts in the areas of Air Quality and Hydrology result from the reduced amount of ground disturbance and building demolition, resulting in fewer impacts from fugitive dust and storm water runoff during construction. The reduced Visual Impact comes from maintaining the PCE building, resulting in less change to the visual character of the area.
These all appear to be very minor advantages when compared to the AIS design, particularly since the Silvergate Substation site is located in an industrial area, and either the GIS or AIS would be replacing a large decommissioned power plant.
While the GIS design avoids relocation of the existing business, according to the record in this proceeding, the existing business is quite willing to relocate. (Yari, supra, Transcript p. 45; Declaration of Kerry J. Lynch, p. 2.)
The potential for additional fugitive dust and storm water runoff is only during construction; the Final EIR finds that operational impacts in these areas would be substantially the same for either design. While the demolition of the PCE building and the construction on its site could have additional impacts in these areas, they are relatively minor in the context of the project, as both the GIS and AIS designs require demolition of the adjacent and much larger Silvergate Power Plant.
Reducing change to the visual character of the area by maintaining the PCE building appears to be a mixed blessing, as the PCE building (apart from a large American flag painted on one wall) does not appear to possess any particular architectural distinction. (DEIR, Figure D.13-2A.)7
One possible alternative not analyzed in the EIR would be a GIS facility, but with the PCE building removed and replaced by landscaping. This would provide a significant visual buffer between the new substation and Harbor Drive, the trolley tracks, and the residential and commercial neighborhood to the east, and would probably also improve water runoff conditions. However, because this alternative was not analyzed in the EIR, we will not consider it here.
Overall, the environmental benefits of the GIS over the AIS are quite small. We must also consider other relevant factors in addition to environmental impacts. Here those factors are primarily the relative cost of each design and its date of operation, especially given the potential reliability issues that arise from the age of the existing Main Street Substation.
According to SDG&E's application, the cost of an AIS is estimated to be $80,600,000. SDG&E has stated that this estimate is no longer accurate, and that the actual cost of an AIS facility is higher, and has requested that the Commission add a 10% contingency to this estimate on any cost cap for the AIS project. (Lynch Declaration, pp. 4-5.) 8 SDG&E estimates that a GIS facility will cost approximately an additional $29,422,000, for a total cost of $110,022,000. (Lynch, supra, p. 4.)
Neither the estimate for the AIS, nor the estimate of the increased cost for a GIS appear to be very precise. SDG&E has stated that the AIS estimate is no longer current, and the GIS estimate has a number of fairly soft elements in it, such as $6,040,000 for "Market Pressure," which is essentially a contingency for unanticipated cost increases. (Yari, supra, Transcript pp. 52-55.) It appears that SDG&E may have presented a "best-case" (or at least a "pretty good-case") scenario for the costs of an AIS, and a "worst-case" scenario for the costs of a GIS.
Using the numbers on the record, all of which were prepared by SDG&E, we should also compare the "worst case" cost of an AIS and the "best case" cost of a GIS. Adding the 10% contingency to the $80,600,000 cost of an AIS results in an estimate of $88,660,000, while subtracting the $6,040,000 for "market pressure" from the cost of a GIS results in an estimate of $103,982,000, for a difference in cost of $15,322,000. This is probably a somewhat more realistic cost differential, given that the GIS estimate is more recent than the AIS estimate. (Yari, supra, Transcript, p. 55.)
While $15,322,000 is significantly less than the $29,422,000 calculated by SDG&E, the GIS design is still quite a bit more expensive than a conventional AIS design. It is not clear that the relatively minimal environmental benefits of the GIS alternative identified in the Final EIR are worth over $15 million.
Another factor we must consider is the date at which a new Silvergate Substation could enter service. According to SDG&E, an AIS facility should be completed and on-line by June 2008, while a GIS facility would not be completed until the first quarter of 2009. (Yari, Transcript, p. 66.) In the larger picture, this difference of roughly six to ten months is not very big, particularly for a facility that is designed to last 30 years or more. However, given the current age and condition of the Main Street Substation, and its growing load, the additional delay could be quite significant. As discussed above, by 2008 SDG&E could be in violation of the reliability criteria of the NERC, the WECC, and the Cal ISO because of the condition and configuration of the Main Street Substation. (Yari, supra, p. 17.)
If the Main Street Substation or the 138 kV lines serving it should fail before the completion of the Silvergate Substation, the consequences could be both serious and expensive. Construction of a GIS facility would increase the risk of such a failure by delaying the completion of the Silvergate Substation at a time when the Main Street Substation is simply not reliable. Accordingly, the difference in the operational dates of the two designs strongly favors an AIS.9
The significant cost and reliability advantages of an AIS clearly outweigh the minor environmental benefits of a GIS. While we must consider the environmental information contained in the EIR, the EIR does not require us to reach a particular outcome; the EIR is primarily an informational document, and does not control our ultimate discretion. (CEQA Guideline 15121; Carmel Valley View, Ltd. v. Board of Supervisors, (1976) 58 Cal. App. 3d 817, 822.) We have considered the information contained in the Final EIR, along with the other evidence in the record, and we find that the AIS design is preferable. We approve SDG&E's request to use an AIS design for the construction of the Silvergate Substation, with a cost cap that allows for a 10% contingency.
7 SDG&E argues that the GIS design has a greater visual impact than the AIS design because the GIS would be enclosed in a solid building, rather than the open-air design of the AIS. (Lynch, supra, p. 2.)
8 SDG&E was directed to update its cost estimates for an AIS facility (Administrative Law Judge (ALJ) Ruling Setting Evidentiary Hearings, p. 2), but did not do so. (Yari, supra, pp. 50-51.)
9 If SDG&E had applied in a timely manner for a CPCN to construct the Silvergate Substation, we could have considered the question of GIS v. AIS based more on their long-term merits, rather than short-term need, as the extra time required to construct a GIS facility would not have mattered. A crisis atmosphere is not conducive to sound policy analysis, and in this case the crisis of Main Street Substation's deterioration was both foreseeable and avoidable.
