What effect will a spaceport in Ka’u have upon the health of people on the Big Island?
Risk Science Associates, a firm in Corte Madera, California, conducted a study of precisely this question. The results are presented in Volume VIII, Appendix A, of the Draft Environmental Impact statement prepared for the state Department of Business, Economic Development, and Tourism.
The study considers the substances released during normal launches aluminum oxide (Al203), released as a particulate, and hydrogen chloride (HCl), released as a gas. It considers the health impact of what are called launch anomalies, including what could be expected to happen if a rocket is destroyed seconds after launch, or if it falls back on the pad and burns. It looks at possible effects of spills of hazardous chemicals on the site or en route to Palima Point.
So-called sensitive receptors: schools, day care centers, hospitals and the like from Pahala to Captain Cook – were identified and a map developed to show these sites and the likelihood of their being subjected to harmful concentrations of the various chemicals emitted under one or another accident or launch scenario. Finally, Risk Science Associates calculated the long-term and short-term risks that someone living or working in the area might, as a result of activities at the spaceport, develop cancer or other diseases.
Poor Pahala
Risk Science Associates concludes that normal launches of the type of rockets proposed for use at the spaceport pose virtually no risk of harm to the nearby population. Almost all of the dangers would arise from accidents, particularly ones involving the hypergolic propellants.
Generally, regulatory agencies regard a risk finding of one excess cancer per million population to be acceptable. Predictions of more than one tend to result in calls for mitigating measures. Non-cancer risks are expressed in a different way. If the probability is 100 percent that one person exposed to a hazard will be injured in some fashion by that exposure, that hazard index is designated as 1.0. When the hazard index is less than one, usually there is thought to be no need for mitigation. When the hazard index exceeds one, usually some thought must be given to reducing the likelihood of injury. Two non-cancer hazard indices were developed in assessing the spaceport risks: one for chronic hazards (for repeated exposure to relatively low concentrations of dangerous chemicals), and one for acute hazards (single exposure to relatively high concentrations of dangerous chemicals).
People living in Pahala would be hardest hit, according to the DEIS study. The acute hazard index for a launch accident is given as 2.0; that for a non-site spill of hydrazine is 10.6, for UDMH it is 2.9, and for nitrogen dioxide (nitrogen tetroxide), it is 7.1. The cancer risk for Pahala is greatly increased by such an event. Roughly one excess cancer per 1,000 population could be expected to occur following a spill of UDMH under worst-case conditions at the spaceport.
Poorer Workers
Workers at the spaceport are at greatest risk. The acute hazard index calculated for a launch accident is 9.1. For an on-site spill, the hazard index range is from 65 to 2,490. According to Risk Science Associates, “exposures where the hazard index exceeds 100, more severe effects such as pulmonary edema could possibly occur in the exposed population.”
The acute hazard index for people within 20 meters of a spill of liquid propellant en route to the spaceport is almost as high. The range for that is from 34 to 510.
Discrepancies
In calculating the health risks associated with fuel shipments to Palima Point, Risk Science Associates relies upon another report in the DEIS – that prepared by ACTA, describing the areas that might need to be evacuated in the event of a transportation accident. (That report is the subject of the [url=/members_archives/archives_more.php?id=1200_0_31_0_C]cover article[/url] in this issue of Environment Hawai`i.)
The scenarios depicted in the ACTA study assume that no more than one container of chemical is spilled, regardless of how many containers were in the shipment. ACTA assumed that the chemical would spill into a pool half a centimeter deep, and that for most chemicals, a pool blanketing time (the amount of time the pool was present on the ground) would be 30 minutes. For accidental spills of a 55-gallon drum of hydrazine and a 300-gallon cylinder of nitrogen tetroxide, ACTA calculated also the area that would be affected assuming a 15-minute blanketing time.
In computing health impacts from transportation spills, Risk Science Associates appears to have ignored the calculations associated with the 30-minute blanketing time and has relied solely on the more limited 15-minute blanketing time scenarios.
Thus, Risk Sciences Associates states: “The highest 60-minute time mean concentrations of hydrazine and [nitrogen tetroxide] are predicted to occur under daytime conditions at a range of 20 meters as a result of a transportation accident at Location I” (Hilo Harbor). The predicted concentration at 20 meters from a spill of nitrogen tetroxide under such circumstances would be 748 parts per million. This agrees with the table in ACTAs study, listing predicted concentrations.
But if one considers the ACTA table showing concentrations of a spill that has a 30-minute blanketing time, the level of nitrogen tetroxide 20 meters downwind of a spill at Hilo Harbor is almost twice that- 1,434 parts per million, or roughly 680 times the short-term public exposure guideline level for nitrogen tetroxide (0.12 ppm).1
Lucky You Live Hawai’i
Many of the accident scenarios analyzed by Risk Science Associates are deemed to be highly improbable. For example, the risk of a launch failure is considered to be so remote as to occur only once during the average 70-year lifetime of a person living nearby. A separate ACTA report – “Impacts of Toxic Spills, Exhaust Gases and Vehicle Accident Gases” – contains a better informed discussion of the probabilities of launch accidents.
That report Volume VI, Appendix D of the DEIS) states:
“To give some perspective on the probability of these types of failures [on-pad aborts and aborts during ascent], the launch histories of the Delta and Atlas vehicles can be examined. During the history of the Delta family of launch vehicles from 1960 to 1990 there have been 12 failures in 201 launch attempts. Of these 12 failures, 4 were related to Stage I or solid rocket motor failures that affect the early phases of a launch. The Atlas family of vehicles experienced 32 failures in 245 launch attempts from 1958 to 1990. Most of these failures occurred prior to 1965. From 1970 to 1990 there were 100 launches and 10 failures. Seven of the ten failures were related to the booster/sustainer stage of the Atlas. Assuming 5 or 6 launches per year from the proposed Hawaiian spaceport, it would not be unreasonable to expect several conflagration-type vehicle failures over a 15-year period.”
Actually, the Office of Space Industry has predicted a much higher use of the spaceport than that assumed by ACTA. In a conceptual plan released in February 1993, OSI projects as many as 10 launches of medium-sized expendable launch vehicles each year (vehicles of the Atlas and Delta size).
The health risk assessment is based upon a failure rate that is not in keeping either with the projected launch rates or the actual failure rate for the vehicles to be used. Revising it to reflect these rates more accurately would almost certainly increase the long-term cancer risks as well as the chronic and acute health hazard indices.
Inescapable Nature
The discussions of accident scenarios and probabilities, while purporting to reflect the “reasonable worst-case” conditions, do not take into account the very real prospect of natural disasters occurring at the spaceport site. These are, however, analyzed at length elsewhere in the DEIS.
A study by ACTA, “Probabilistic Analysis of Inundation by lava Flow from Volcanic Activity” (attached to the geologic report that appears as Volume III, Appendix A), found that over the predicted 50-year lifetime of a spaceport, flows from the summit and southwest rift zones of Mauna Loa volcano can be expected to empty into the Palima Point catchment area at least once. ACTA found that there was a 39 percent probability of a flow from Kilauea overrunning the site during that same period. “This estimate maybe somewhat low, however,” it noted, “because it has been 170 years since the last eruption on the lower SWRZ [southwest rift zone] and because there has been significant intrusive activity… on the lower SWRZ… These observations suggest that an eruption from the SWRZ of Kilauea volcano may be very likely during the next 50 years.” The area of Palima Point most likely to he affected would be the coastal strip, along which the launch pads are proposed to be built.
Earthquake
The proposed spaceport would be built in an area that is but a few kilometers from the epicenter of Hawai’i’s greatest earthquake -the devastating earthquake of Ka’u that hit in 1868, and whose magnitude is estimated to have been between 7.5 and 8 on the Richter scale. As is explained in the study of seismic risks, also prepared by ACTA, “the installation of a rocket launch facility at the Palima Point site would alter the seismic risk to the surrounding population and environment in two ways: It would introduce hazardous materials, primarily in the form of rocket propellants, to the region, which if not adequately contained during an earthquake could pose a risk to both the population and the environment. The installation would also increase the population of the area, exposing more people to the consequences of a hazardous materials release.”
Buildings associated with spaceport operations can be constructed to withstand earthquakes. With careful attention and design, it may be possible to design storage areas for hazardous chemicals capable of standing up to major earthquakes. According to ACTA, however, it will be extremely challenging to design launch pads so that rockets in position for launch will be insulated from seismic impacts. “Intuition and experience suggest that the most likely ‘worst-ease scenario’ would be the occurrence of a major earthquake while a launch vehicle is fully loaded on the pad prior to lift-off,” the ACTA study states. “This scenario involves the most vulnerable configuration of propellant storage in that large quantities of both fuel and oxidizer are in close proximity to each other, the launch vehicle has minimal lateral support to withstand earthquake forces and would be exposed to puncture via toppling or slamming against adjacent support towers.”
The report on seismic risks, unlike other descriptions of possible accident scenarios, mentions the prospect of a disaster causing more than one chemical to be released to the atmosphere. “The three principle consequences of propellant releases are leaks or spills, fires and explosions. Any one of the three may result in the other two.”
According to ACTA, “magnitude 6.6-7.5 earthquake is considered likely to occur during the 50-year life of the facility.”
Blasts and Booms
The health risk assessment does not consider what elsewhere in the DEIS are described as “explosive overpressure hazards.” As stated in the introduction to an ACTA report analyzing these hazards (Appendix F to Volume VII of the DEIS), “An accident on the pad or the fail-back of a rocket shortly after lift off under some circumstances can produce a high-yield explosion. This explosion will generate a shockwave that will propagate through the atmosphere and affect areas that may be a significant distance away from the launch pad. Although these shock waves attenuate with distance from the source, certain atmospheric conditions will either decrease the rate of attenuation or cause amplification because of the convergence of the acoustic ray paths… If these atmospheric conditions lead to overpressures of .05 to .50 [pounds per square inch] in populated areas, significant glass breakage can occur. The glass breakage can also lead to lacerations to people near windows. The higher the overpressure, the more likely the casualties.”
A table in this report shows that the maximum overpressure that would occur in the explosion of a Delta rocket on the pad would be 0.178 psi at Pahala. At nearby Punalu’u, the pressure would be 0.174 psi. At Na’alehu, the expected pressure would be 0.046 psi just below the 0.05 psi threshold above which breakage is expected. Under certain conditions, the report states, “The probability of breakage of type C windows [windows whose area is between 10 and 40 square feet] is approximately one fifth. In other words, under the above described conditions, for every 10 windows of type C at Pahala, two of them will break and perhaps cause an injury due to flying glass.”2
The health risk assessment also did not consider the impacts of noise, although an ACTA “Acoustic Environment Assessment” reports that noise levels can be extremely high – up to 109 decibels for people in Pahala and Punalu’u, under certain atmospheric conditions. “Such levels can be expected to cause window rattle but are not expected to cause any damage to windows in normal condition at these locations,” the acoustic report states. “Some limited amount of window rattling may occur for the more distant areas such as Hilo or the Mauna Kea Observatory,” where maximum sound levels could go as high as 90 decibels under adverse weather conditions.
Finally, there are the sonic booms that predictably accompany launches of vehicles such as the Atlas and Delta rockets. “A Brief Survey of Sonic Boom Consequences” prepared by ACTA (Volume VII, Appendix E of the DEIS) shows that the area around Kapoho will be most heavily affected by these. “There is a possibility of breaking windows and of damaging plaster walls,” the report states. Estimated pressures for Atlas and Delta rockets go as high as 0.052 pounds per square inch. At that pressure, some window breakage can be expected. ACTA estimates the breakage maybe as low as three panes per 100,000 panes subjected to the boom, or as high as three panes per 1,000.
Poisoned Waters
With surface water and catchment systems providing most of the drinking water sources in Ka’u, the possibility that water will be contaminated by spaceport operations is considered in the health risk assessment. Under normal launch conditions, the report says, there will be negligible long-term cancer risk of drinking water where concentrations of hydrazine are elevated as a result of a mid-air explosion.
The non-cancer hazard is substantially higher, according to the report. The “non-cancer hazard index” from drinking water contaminated with hydrazine as might occur if it rains shortly after the “deflagration event” -is put at 2.0, which means it is highly likely that some people would be affected. Ingestion of hydrazine can cause liver damage, nausea, anorexia, and convulsions.
1 Reading the Risk Science Associates report, one might think there are other discrepancies. For example, RSA states that it assumed that the spill scenarios it considered involved the contents of two 55-gallon drums being emptied (ACTA assumed just one), and that the pool depth (which determines evaporation rate) would be 1 centimeter; ACTA assumed a depth of 0.5 centimeters. A statement of “Revised Assumptions” appears at the beginning of Volume VIII, however, correcting the assumptions regarding pool depth and number of containers ruptured to be in line with the assumptions made by ACTA. “The analysis is not affected by these changes,” the note states.
2 It may be worth mentioning that the Office of Space Industry was not pleased with this ACTA report. Its appearance in the Draft EIS (as Volume VIII, Appendix F) is followed immediately by a response prepared by the Office of Space Industry (Appendix G). The OSI response takes exception to several of the assumptions contained in ACTA’s report. In summary, OSI states: “It appears that ACTA’s methodology may be putting a stronger burden upon the proposal Hawai`i spaceport than that currently imposed at the national ranges (for both commercial and military launches)… New facilities must not be held to more rigorous standards than those used to successfully and safely conduct daily operations at the existing national ranges.”
Volume 4, Number 2 August 1993