In Town of Weymouth v. Mass. Dept. of Env. Prot., 2020 WL 2904672 (1st Cir. 2020), the First Circuit vacated and remanded an air permit issued by the Massachusetts Department of Environmental Protection (“DEP”) to Algonquin Gas Transmission, LLC (“Algonquin”) for its proposed compressor station in Weymouth Massachusetts. Although the permit is being sent back to DEP for further review, the Court narrowed the issues on remand by ruling on the merits of each of the petitioners’ claims, finding in DEP’s favor on all but one. The only issue for DEP on remand is whether an electric motor is the Best Available Control Technology (“BACT”). The Court’s opinion provides important guidance for pipeline companies that are seeking to expand capacity on their systems through compression.
Algonquin’s proposed compressor station, which received a certificate of public convenience and necessity from the Federal Energy Regulatory Commission (“FERC”), was to include a natural gas-fired turbine to compress gas along Algonquin’s pipeline system. Prior to constructing the compressor station, Algonquin was required to obtain all necessary federal authorizations, including permits under the Clean Air Act. To that end, Algonquin applied to the DEP for air permits in October 2015. After some litigation to compel DEP to make a decision, and settlement, DEP issued the permit in January 2019. An administrative appeal was then filed by the Town of Weymouth, other nearby municipalities, residents of the local municipalities, and state and local officials (“Petitioners”) based on claims that the permit violated Massachusetts law. The administrative appeal resulted in the affirmance of the permit and the Petitioners filed an appeal to the First Circuit.
DEP’s BACT analysis for an Electric Motor
The main challenge to Algonquin’s air permit was to DEP’s BACT analysis as to Nitrogen Oxide (NOx). DEP has adopted the U.S. Environmental Protection Agency’s (“USEPA”) five-step “top-down” approach to determining the BACT. These steps are: (1) identify all control technologies; (2) eliminate technically infeasible options; (3) rank the remaining control technologies by control effectiveness; (4) evaluate the most effective controls and document results; and (5) select the BACT. After applying this test within its air permit application, Algonquin ultimately concluded that the SoLoNOx turbine was the BACT for NOx, and DEP agreed. However, the Petitioners claimed that DEP’s BACT analysis was deficient because (1) it excluded the use of an electric motor driven compressor and (2) wrongly determined that a Dry Low NOx turbine plus selective catalytic reduction (“SCR”) was not cost feasible.
Generally, electric driven compressors are powered by electricity and allow pipeline companies to add compression to their pipeline systems without having to burn natural gas, thereby eliminating NOx emissions. The downside to these types of compressors is that they require substantial investment, including the construction of an electric substation and electric transmission lines to bring power to the compressor station site. Further, the construction of these facilities can have additional environmental impacts and impacts to landowners. There are also certain operational constraints associated with electric driven compressors in that they cannot operate in the event the electric grid is down, such as during severe weather events, whereas gas-fired compressors can still operate.
In its application, Algonquin rejected the electric driven compressor, in part, due to the high cost of upgrading the power infrastructure and the fact that FERC concluded that it did not provide a significant environmental advantage over the proposed gas-fired turbine. As noted by the Court, DEP accepted Algonquin’s conclusion and relied on FERC’s finding without performing its own independent BACT analysis. However, during the administrative appeal, Algonquin and DEP argued that the electric motor was appropriately excluded under step 1 of the BACT analysis as it would “redefine the source”. Algonquin further submitted testimony that the cost of the electric motor and the infrastructure upgrades attendant with that alternative would cost between $9 million and $12 million. The officer that presided over the administrative appeal did not rule on whether the electric motor would redefine the source, but instead relied on the cost estimate provided by Algonquin in holding that the electric motor would not be cost feasible and excluded the electric motor under step 4 of the BACT analysis.
On appeal to the First Circuit, Petitioners claimed that the testimony of Algonquin’s witness was not substantial evidence sufficient for the Court to uphold the agency’s decision; (2) that neither DEP nor the presiding officer appropriately performed step 4 of the BACT analysis pursuant to the DEP’s guidance document; and (3) that the presiding officer only raised the step 4 analysis during the administrative appeal without providing other parties with an opportunity to comment on the analysis.
The Court held that the evidence before the presiding officer was not insubstantial as a matter of law as the testimony from Algonquin’s witness as to the cost of the electric motor was not “irrational” and Petitioners provided no evidence contradicting Algonquin’s estimate. However, the Court held that DEP’s exclusion of the electric motor on cost effectiveness grounds was arbitrary and capricious.
Interestingly, the Court went so far as to attempt to calculate the per ton cost in accordance with DEP’s manual. However, it could not determine the average or incremental costs associated with using an electric motor based on the little information it could glean from the record. The Court held that it was not enough for Algonquin and DEP to simply say that the capital costs for the electric motor were far greater than other cost-infeasible control technologies and, therefore, obviously more infeasible. The Court concluded that DEP’s BACT guidance requires a cost-effectiveness analysis before the electric motor could have been excluded under step 4 and the Court could not find harmless error in DEP’s failure to follow its own guidance or provide enough of an analysis that would eliminate the electric motor as not cost-effective.
DEP’s Exclusion of Selective Catalytic Reduction
In addition to challenging DEP’s exclusion of the electric motor, Petitioners also challenged the exclusion of selective catalytic reduction (“SCR”). Unlike the use of the electric motor, which the Court identified as a process-control technology, SCR is an add-on technology used to reduce NOx emissions. In its permit application, Algonquin identified SCR as the most effective technology but excluded it as not being cost feasible under step 4 of the BACT analysis. DEP’s stated range for cost-effective technologies is $11,000-$13,000 per ton. Under Algonquin’s analysis, SCR’s cost-effectiveness was determined to be $41,541 per ton of NOx removed, based on a baseline of 9 parts per million (ppm).
The Petitioners argued that Algonquin should have used a baseline emissions rate of 25 ppm, to reflect older models of Dry Low NOx turbines, which would have resulted in a lower per-ton cost. However, as the Court noted, even assuming Petitioners were correct that DEP should have used the higher baseline rate, the cost effectiveness would be $14,483 per ton, which is greater than DEP’s $13,000 threshold. Petitioners claimed that the $13,000 was established by DEP in 1990 and should be adjusted for inflation, thereby making SCR cost feasible. The Court, however, found that it did not need to address Petitioners inflation argument as DEP used the appropriate baseline rate pursuant to its guidance. Under DEP’s guidance, in determining the cost effectiveness of adding post process emissions controls (in this case, SCR) to certain inherently lower polluting processes (here, SoLoNOx), baseline emissions may be assumed to be the emissions from the lower polluting process itself (SoLoNOx). Accordingly, the 9-ppm emissions rate from SoLoNOx was the appropriate baseline.
The Court also upheld DEP’s review of the air permit application in light of Massachusetts’ Environmental Justice Policy and noise regulations. The Court was not persuaded that DEP was required to provide additional process required under Massachusetts’ EJ Policy because there was no evidence that Algonquin’s proposed compressor station would exceed the emissions thresholds that would trigger the EJ Policy. The Court also rejected Petitioners’ claims that a recent decision by the U.S. Court of Appeals for the Fourth Circuit in Friends of Buckingham v. State Air Pollution Control Bd., 947 F.3d 68 (4th Cir. 2020) required DEP to do more as that decision pertained to Virginia’s EJ Policy, not Massachusetts’ EJ Policy. The Court further held that DEP did not violate noise regulations but required Algonquin to mitigate for noise impacts associated with construction and operation of the compressor station.
Lastly, the Court upheld DEP’s reliance on manufacturer guarantees for equipment emissions. The Court held that even if there were no evidence of the manufacture’s guarantee in the record, Petitioners could not show that a manufacturer guarantee is required under Massachusetts law. The Court noted that if the turbine fails to operate as required by the permit, Algonquin would be subject to fines and required to limit its operation of the compressor station to comply with the terms of the permit.
The Court ultimately vacated and remanded the permit based on DEP’s failure to properly perform the BACT analysis for the electric motor. The Court is allowing DEP to reopen the record to obtain further information on the costs of the electric motor. Algonquin would also be allowed to further its argument that the electric motor would redefine the source. It will be interesting to see how DEP addresses Algonquin’s argument, especially in light of the Fourth Circuit’s decision in Friends of Buckingham.
The Court’s decision provides helpful guidance for prospective air permit applicants, especially in regards to the BACT analysis. It is not enough to set out the total costs of the control technology. An applicant needs to connect all the dots and do the math to establish that the technology is not cost-effective. Furthermore, depending on when cost effectiveness thresholds were established, it may be necessary for applicants to adjust those thresholds for inflation to further establish that a technology is not cost effective.