Follow-Up Request for a Supplemental EIS on the proposed PSE LNG plant at the Port of Tacoma

Here’s what was just sent (7/19/17 at 11:19pm PST) to the City of Tacoma and State of WA (Dept of Ecology):


Dear City of Tacoma and WA Department of Ecology,

Tomorrow will make right at two weeks since I submitted the request for a supplemental EIS on 7/6/17 and I’ve yet to hear a response (while PSE continues to drill new pipes across Fife and the Tideflats).

I’m hopeful that you will soon agree to conduct the Supplemental EIS addressing the issues in my 7/6 letter (which was revised 7/7). You can see all of the attachments and the original Supp-EIS request letter on this link:

I’m also hopeful that the Port Commission voices a public commitment to support a supplemental EIS at their meeting tomorrow.

In addition to my letter and its attachments, you should also be aware of other recent EIS’s that have been conducted that incorporate a full safety analysis as opposed to the abbreviated and incomplete one conducted for the City of Tacoma, including:

1) Oregon’s Jordan Cove  conducted by the Federal Energy Regulatory Commission, which stated that:

“The principal hazards associated with the substances involved in the liquefaction, storage, and vaporization of LNG result from cryogenic and flashing liquid releases, flammable and toxic vapor dispersion, vapor cloud ignition, pool fires, BLEVEs, and overpressures.”

4.13.4 LNG Facility Siting Requirements.

Although those are the principal hazards, even these principal hazards were not addressed (or definitely not in the manner necessary to advise the public and have the review organization make an informed decision).

The Jordan Cove EIS also:

a. Required that the LNG facility adopt federal safety standards to address thermal radiation (even though it will be located in a rural area and not a major city):

the LNG terminal would meet the federal safety regulations regarding the thermal radiation and flammable vapor dispersion exclusion zones and appropriate design
Jordan Cove Energy and Final EIS Pacific Connector Gas Pipeline Project standards, and Pacific Connector’s natural gas facilities would also be designed, constructed, and operated in accordance with DOT safety standards

Executive Summary ES-18. See also “Thermal Radiation Analysis” found on page 4-1018 of the Jordan Cove Final EIS.

In Jordan Cove’s case, an exclusion zone of 985 feet was found. [Here] that would put the new [Port] fire station within range, as well as other facilities which have the capacity to house more than 50 people at one time (including the entrance to the TOTE facilities and it would also appear to include the tanker trunks and part of TARGA’s tank farm). This is obviously a very critical safety assessment that must be conducted by an independent lab/safety engineer pursuant to a supplemental EIS.

As you’ll read from an EIS that more fully address the safety risks, exclusion zones are defined those zones around an LNG facility that would be exposed to thermal radiation (something that the LNG industry engineers that assisted the city of Tacoma simply ignored, as many LNG industry representatives do). (See 4-897)

Thus, in addition to concrete “capturing” vessels that appear to well exceed anything PSE has proposed, the Jordan Cove facility was to include 8 foot, 12 foot, 20 foot and 40 foot high impermeable vapor barriers along the property line to try and confine vapor clouds and try and limit the extent of the vapor dispersion zone. (see 4-996 to 997) In addition to many safeguards that the PSE LNG plant was not required to have, the Freeport LNG facility was also to have vapor barriers (See Freeport Final EIS, at pg 4-180)

b. (The Jordan Cove EIS) included 443 letters with comments on the D-EIS (which should be compared to the Tacoma D-EIS process which did not openly or sufficiently engage the public); and

c. (The Jordan Cove EIS) included an analysis of real, substantial and actual “alternatives” to be considered that then resulted in required mitigation strategies (something that the City of Tacoma’s EIS didn’t include, other than “assurances” that PSE would do their best to mitigate in light of there being no alternatives that were acceptable to PSE);

2) A recent EIS was also conducted on the Freeport (TX) LNG Liquefaction Project

This EIS makes clear that there are safety aspects NOT CONSIDERED in the City of Tacoma’s very brief safety analysis. One can just compare page 4-142 onward of the FERC Final EIS for Freeport to that generated by the City of Tacoma to see the incredible difference in amount of time and resources and explanation that went into one but not the other. The Freeport EIS reads:

The transportation of natural gas by pipeline involves some risk to the public in the event of an accident and subsequent release of gas. The pipeline facilities as part of Liquefaction Project are identified in table 2.1.3-1 and include the BOG pipeline and interconnects. In addition to the natural gas pipelines, there would be a water pipeline, nitrogen pipeline, and a nonjurisdictional NGL pipeline.
In regards to natural gas pipelines, the greatest hazard is a fire or explosion following a major pipeline rupture. Methane (CH4), the primary component of natural gas, is colorless, odorless, and tasteless. It is not toxic, but is classified as a simple asphyxiate, possessing a slight inhalation hazard. If breathed in high concentration, oxygen deficiency can result in serious injury or death.” [4-142 to 143]

“4.10.2 Hazards
The principal hazards associated with the substances involved in the liquefaction, storage and vaporization of LNG result from loss of containment, vapor dispersion characteristics, flammability, and the ability to produce damaging overpressures. A loss of the containment provided by storage tanks or process piping would result in the formation of flammable vapor near the release location, as well as the potential for nearby pooled liquid. Releases occurring in the presence of an ignition source would most likely result in a fire located at the vapor source. A spill without ignition would form a vapor cloud that would travel with the prevailing wind until it either dispersed below the flammable limits or encountered an ignition source. In some instances, ignition of a vapor cloud may produce damaging overpressures. The dispersion of toxic components would also be a hazard associated with substances at the Pretreatment Plant. These hazards are described in more detail below.”

“ Flammable Vapor Dispersion
In the event of a loss of containment, LNG, refrigerants (including ethylene and propane) and NGLs would create vapor when released from storage or process facilities. Depending on the size of the release, a liquid pool may also form and vaporize. Additional vaporization would result from exposure to ambient heat sources, such as water or soil. When released from a containment vessel or transfer system, LNG would produce about 620 standard cubic feet of natural gas for each cubic foot of liquid. Each cubic foot of refrigerants or NGLs would generally produce a similar or smaller volume of vapor upon release than would be generated by LNG.
If the loss of containment does not result in immediate ignition of the LNG, refrigerant, or NGL vapors, the vapor cloud would travel with the prevailing wind until it either encountered an ignition source or dispersed below its flammable limits.
An LNG vapor cloud would initially sink to the ground due to the cold temperature of the vapor. As an LNG vapor cloud disperses downwind and mixes with the warm surrounding air, the LNG vapor cloud may become buoyant. The LNG vapor cloud would not typically be warm, or buoyant, enough to lift off from the ground before the LNG vapor cloud becomes too diluted to be flammable. As an ethylene vapor cloud disperses downwind and mixes with the warm surrounding air, the ethylene vapor would become neutrally buoyant. However, a dispersing propane vapor cloud would remain denser than the surrounding air, even after warming to ambient temperatures. The buoyancy of a NGLs vapor cloud would depend on its composition, which would vary, and this vapor could be either positively or negatively buoyant. As a result, estimating the dispersion of the vapor cloud is an important step in addressing potential hazards and is discussed in section 4.10.5 for the facilities at the terminal and in section 4.10.6 for the Pretreatment Plant. Vapor Cloud Ignition
The flammability of a vapor cloud is dependent on the concentration of the vapor when mixed with the surrounding air. In general, higher concentrations within the vapor cloud would exist near the spill, and lower concentrations would exist near the edge of the cloud as it disperses downwind. Mixtures occurring between the lower flammability limit (LFL) and the upper flammability limit (UFL) could be ignited. Concentrations above the UFL or below the LFL would not ignite.
The LFL and UFL for methane are between 5 -percent-volume and 15 percent-volume in air, respectively. Propane has a narrower flammability range, but has a lower LFL of approximately 2.1 percent-volume and a UFL of 9.5 percent-volume in air, respectively. Ethylene has a much wider flammability range and a lower LFL of approximately 2.7 percent-volume and a UFL of 36 percent-volume in air. Mixed refrigerant would have a UFL and LFL based on the amount of LNG, ethylene, and propane it contains, which would vary throughout the process. NGLs would have similar UFLs and LFLs based on the amounts of heavier hydrocarbons it contains, which would also vary.
If the flammable portion of a vapor cloud encounters an ignition source, a flame would propagate through the flammable portions of the cloud. In most circumstances, the flame would be driven by the heat it generates, a process known as a deflagration. A methane vapor cloud deflagration in an uncongested and unconfined area travels at slower speeds and does not produce significant pressure waves. Confined and congested methane vapor clouds may produce higher flame speeds and overpressures, and are discussed later in section

Once the flammable portion of a vapor cloud has encountered an ignition source, a deflagration may propagate back to the spill site if the vapor concentration along this path is sufficiently high to support the combustion process. When the flame reaches vapor concentrations above the UFL, the deflagration could transition to a fireball and result in a pool or jet fire back at the spill source. A fireball would occur near the source of the release and would be of a relatively short duration compared to an ensuing jet or pool fire. Radiant heat modeling for pool fires at the terminal site is discussed in section Radiant heat modeling for pool fires at the Pretreatment Plant is discussed in section

The extent of the affected area and the severity of the impacts on objects either within an ignited cloud or in the vicinity of a pool fire would primarily be dependent on the quantity and duration of the initial release, the surrounding terrain, and the environmental conditions present during the dispersion of the cloud. A vapor cloud fire can ignite combustible materials within the cloud and can also cause severe burns and death. Fires may also cause failures of nearby storage vessels, piping, and equipment. The failure of a pressurized vessel could cause fragments of material to fly through the air at high velocities, posing damage to surrounding structures and a hazard for operating staff, emergency personnel, or other individuals in proximity to the event. In addition, failure of a pressurized vessel when the liquid is at a temperature significantly above its normal boiling point could result in a boiling-liquid-expanding-vapor explosion (BLEVE). BLEVEs of flammable liquids can produce overpressures and a subsequent fireball when the superheated liquid rapidly changes from a liquid to a vapor upon the release from the vessel. This concern is addressed in section for the pressurized propane and ethylene storage tanks for the Liquefaction Plant. The NGLs at the Pretreatment Plant would not be stored in pressurized tanks. Atmospheric storage tanks, such as those existing and approved for LNG storage at the terminal, are unlikely to BLEVE due to the smaller difference between their design pressure and ambient pressure.”

“ Overpressures
If the deflagration in a flammable vapor cloud accelerates to a sufficiently high rate of speed, pressure waves that can cause damage would be generated. As a deflagration accelerates to super-sonic speeds, larger pressure waves are produced, and a shock wave is created. This shock wave, rather than the heat, would begin to drive the flame, resulting in a detonation. Deflagrations or detonations are generally characterized as “explosions” as the rapid movement of the flame and pressure waves associated with them cause additional damage beyond that from heat. The amount of damage an explosion causes is dependent on the amount that the produced pressure wave is above atmospheric pressure (i.e., an overpressure) and its duration (i.e., pulse). For example, a 1 pound per square inch (psi) overpressure, often cited as a safety limit in U.S. regulations, is associated with shattering glass with glass fragments traveling with velocities high enough to lacerate skin.
Flame speeds and overpressures are primarily dependent on the reactivity of the fuel, the ignition strength and location, the degree of congestion and confinement of the area occupied by the vapor cloud, and the flame travel distance.
The potential for unconfined LNG vapor cloud detonations was investigated by the USCG in the late 1970s at the Naval Weapons Center in China Lake, California. Using methane, the primary component of natural gas, several experiments were conducted to determine whether unconfined LNG vapor clouds would detonate. Unconfined methane vapor clouds ignited with low-energy ignition sources (13.5 joules), produced flame speeds ranging from 12 to 20 mph. These flame speeds are much lower than the flame speeds associated with a deflagration with damaging overpressures or a detonation.

To examine the potential for detonation of an unconfined natural gas cloud containing heavier hydrocarbons that are more reactive, such as ethane and propane, the USCG conducted further tests on ambient-temperature fuel mixtures of methane-ethane and methane-propane. The tests indicated that the addition of heavier hydrocarbons influenced the tendency of an unconfined natural gas vapor cloud to detonate. Natural gas with greater amounts of heavier hydrocarbons would be more sensitive to detonation.

Although it has been possible to produce damaging overpressures and detonations of unconfined LNG vapor clouds, the LNG proposed for liquefaction by this project would have lower ethane and propane concentrations than those that resulted in damaging overpressures and detonations.

The substantial amount of explosives needed to create the shock initiation during the limited range of necessary vapor-air concentrations also renders the possibility of detonation of unconfined LNG vapors as unrealistic. Ignition of a confined LNG vapor cloud could result in higher overpressures. In order to prevent such an occurrence, measures are taken to mitigate LNG vapor dispersion into confined areas, such as buildings, and also the potential for ignition inside them. In general, the primary hazards to the public from an LNG spill that disperses to an unconfined area, either on land or water, would be from dispersion of the flammable vapors or from radiant heat generated by a pool fire, as discussed in the previous sections In comparison with LNG vapor clouds, there is a higher potential for unconfined propane to produce damaging overpressures, and an even higher potential for unconfined ethylene vapor clouds to produce damaging overpressures. Unconfined ethylene vapor clouds also have the potential to transition to a detonation much more readily than propane. This has been shown in multiple experiments conducted by the Explosion Research Cooperative to develop predictive blast wave models for low, medium, and high reactivity fuels and varying degrees of congestion and confinement (Pierorazio, 2005). The experiments used methane, propane, and ethylene, as the respective low, medium, and high reactivity fuels. In addition, the tests showed that if methane, propane, or ethylene is ignited within a confined space, such as in a building, they all have the potential to produce damaging overpressures. The NGLs process streams at the Pretreatment Plant would contain similar or heavier hydrocarbon components. Therefore, a potential exists for these process streams to produce unconfined vapor clouds that could produce damaging overpressures in the event of a release.

These overpressure hazards are discussed in section for the facilities at the terminal and in section for the Pretreatment Plant.”

Obviously, there is a missing safety and mitigation analysis from the City of Tacoma’s Final EIS…since the conclusions are far far different from those prepared by non-industry experts (such as governmental agencies having the resources to conduct an independent report). And, the Freeport TX EIS cites the Algerian and Plymouth WA accidents and evacuations (which the City of Tacoma’s EIS failed to address).

And, I don’t recall an analysis of vapor clouds from Other Hazardous fluids, such as methane, propane and ethylene (as discussed on page 4-1012 of the Jordan Cove EIS:

And, of course, all of this is also crucially important because PSE has time and time again told the public (which was joined by the City of Tacoma) that there is no danger from the LNG facility and that any accident would be limited to the property confines (something that even the Jordan Cove EIS admits is a very real possibility, even absent an intentional attack – See. at 4-1006

Until the City of Tacoma and the State of Washington (Dept of Ecology) agree to look into the safety of the citizens of Tacoma who are legitimately at risk by the proposed plant, I will continue to provide copies of all of the Final EIS’s that have been considered recently and that more adequately address the safety risk (something that the City of Tacoma Final EIS is clearly lacking in).

It’s one thing to not consider these risks (which the Final EIS is guilty of) versus considering them with alternatives and mitigating strategies to limit the risks (which these aforementioned EIS’s do).

-Noah Davis


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s