Best Practices for General Aviation Aircraft Fuel-Tank Sampling (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

AIRPORT COOPERATIVE RESEARCH PROGRAM Sponsored by the Federal Aviation Administration Responsible Senior Program Officer: Theresia H. Schatz July 2014 C O N T E N T S Introduction, 1 Objective 1—Quantity Estimate, 1 Discarded Fuel—Fate, Environmental Implications, and Applicable Regulations, 4 Current State of the Practice, 7 Objective 2—Best Practices, 9 Conclusions and Suggestions, 14 References, 15 • Objective 2—Develop a variety of airport and aviation best practices of aircraft fuel-tank sampling to pre- vent contaminants from entering the stormwater runoff system. Objective 1 was accomplished by conduct- ing pilot surveys, field-based research, and mathematical calculations and Objective 2 was accomplished by performing literature reviews, inquiries and field testing, and preparing this Digest. Given that the best practices presented in this Digest are aviation-based, the intended audience for this document is GA pilots, airport managers, flight schools, and other aviation personnel and organizations. These entities have a role or vested interest in (a) proper fuel management and disposal and (b) the potential consequences if not these are not conducted properly. OBJECTIVE 1—QUANTITY ESTIMATE The Objective 1 quantity estimate was accomplished by collecting raw data from pilot surveys and field-based observa- tions, applying various assumptions, and conducting mathematical calculations to INTRODUCTION As part of a typical general aviation (GA) internal combustion engine preflight inspec- tion, the pilot normally takes fuel samples from fuel system drains (a.k.a. “sumps”) and visually inspects them for color and contam- ination (e.g., water, particulate). Although these fuel samples can be returned to the fuel tank if they are uncontaminated, or disposed of in an approved container if contamination is found, it is generally acknowledged that many pilots simply discard the fuel samples to the ground surface regardless of whether contamination was identified or not. This disposal practice is inconsistent with many airport pollution prevention procedures and environmental regulations and can lead to stormwater runoff contamination and air pollution. The Airport Cooperative Research Pro- gram (ACRP) commissioned this project to evaluate this disposal practice, and spe- cifically, to accomplish the following two objectives: • Objective 1—Estimate the discard amounts from fuel testing samples that are entering the stormwater run- off system. BEST PRACTICES FOR GENERAL AVIATION AIRCRAFT FUEL-TANK SAMPLING This report presents the results of ACRP Project 11-02, Task 22, “Best Practices for General Aviation Aircraft Fuel-Tank Sampling.” The research was conducted by a team lead by Hagerty Environmental, LLC. The Digest summarizes the research and presents several best practices for general aviation fuel-tank sampling. The Principal Investigator (PI) for this research project was Paul Hagerty. Contribution to the research and review process was provided by Carmine Balascio, University of Delaware, and Jonathan Martin, New Garden Township, Pennsylvania. Research Results Digest 21

2arrive at the estimate. The following sections pro- vide an overview of the process and the resulting quantity estimate. Data Collection Methodology Data were collected by conducting an electronic survey of pilots’ typical disposal practices (1). The survey was disseminated via various aviation networks (e.g., Facebook, LinkedIn, Aircraft Own- ers and Pilots Association [AOPA] website forum, pilot-based email lists) and consisted of the follow- ing 11 questions: 1. Do you currently pilot an aircraft with a reciprocating (i.e., piston) engine? 2. How many fuel system sample locations (a.k.a. sumps) does the primary aircraft you fly have (including cowl drains, gascolators, etc.)? 3. Does the primary aircraft you fly have a sump/strainer location which is remotely operated (e.g., inside the top of cowl or in cockpit) that makes sample collection diffi- cult or impractical? 4. Select the response below which best describes your “typical” fuel sampling/sumping procedures? 5. When collecting fuel samples for visual inspec- tion, what type of fuel sampling device do you primarily use? 6. When collecting fuel samples for visual inspec- tion, how much of a sample do you normally collect from each sump location? 7. What do you do with the fuel sample(s) once they have been collected and inspected? 8. How often have you experienced significant contamination in your fuel samples (e.g., water, excess particulate, etc.)? 9. Where did you learn your current fuel testing/ sumping procedures? 10. When did you receive your primary flight training? 11. How many hours of flying do you average per year (reciprocating only)? The survey was available for 1 month from Sep- tember 16 through October 16, 2013. It was com- pleted by 146 respondents. The size of this dataset was considered adequate for two reasons. First, the PI periodically reviewed the survey data over the 1-month survey period and noted that the responses were fairly consistent at the onset of the survey as compared with the survey close date. Second, field- based research was conducted at three GA airports (discussed in the Current State of the Practice section) and corroborated the range and general distribution of the responses from the pilot survey. As an exam- ple of the range of disposal behaviors received from the survey, Figure 1 presents a graphical summary of the responses from survey question 7. Calculation Methodology Generally speaking, the fuel discard estimation was carried out by multiplying the average fuel dis- posal quantity per flight (from the survey data) by the 35.7% 18.8% 25.9% 3.6% 16.1% Queson 7. What do you do with the fuel samples once they have been collected and inspected? Discard all samples to the surface regardless of visible contamina on (35.7%) Discard sample to the ground if visibly contaminated; pour back into the tanks if no contamina on is observed (18.8%) Pour back into the fuel tanks using a fuel straining device (e.g., mesh screen) (25.9%) Pour only the contaminated samples into a ground-based container (e.g., gas can, bucket, etc.) (3.6%) Pour all samples into a ground- based container (e.g., gas can, bucket, etc.) (16.1%) Figure 1 Data summary from survey question 7.

3number of GA operations per year (from Federal Avi- ation Administration [FAA] database) (2). Numer- ous assumptions were made to adjust the survey data and the FAA database data. These assumptions and adjustments are summarized below. • Step 1—Average sump locations per aircraft (from survey results)—3.479 (“A”). • Step 2—Average quantity of a full sump sample (takes into account the respondents’ specific type of samplers used)—42.595 milli- liters (mL) (“B”). • Step 3—Adjusted per-sample quantity based on how full the average pilot fills the sampler (~45% full)—0.45 × B = 19.306 mL (“C”). • Step 4—Average sample quantity per flight— A × C = 67.166 mL (“D”). • Step 5—Adjust per-flight quantity based on the following: Step 5A—Reduce based on pilots disposal habits of returning sample to fuel tanks or placing in disposal containers—0.368 (“E”). Step 5B—Increase based on weighted average of some aircraft that have a re- mote sump location which cannot be col- lected—5.36 mL per flight (“F”). • Average Per-Flight Discard—(D × E) + F = 30.1 mL per flight (“G”). FAA data (2) were reviewed for GA operations in the United States to determine the total number of GA operations per year. This number was then reduced in a stepwise fashion for various known factors as summarized below. • Step 6—Start with raw “operations” data (81,529,543 operations) and apply a series of reduction factors as follows: – Apply 0.5 factor since operations data in- clude take-offs and landings. – Apply 0.5 factor since many take-offs are touch-and-goes for flight training, which do not result in a sumping/sampling event (see Conclusions and Suggestions section for sensitivity analysis). – Apply factor from survey (0.811) to account for pilots who either do not sump upon refueling or do not sump before flight. – Apply factor from FAA data (0.948) for those GA operations that are jet engines. – Total annual GA operations which result in fuel sample discard to ground = 15,680,190. • Step 7—Estimate the total quantity of fuel dis- carded to the ground surface. – Adjusted per-aircraft sump quantity times adjusted annual GA operations = 471,882,554 mL. – Converted to gallons  124,672 gallons per year. Quantity Estimate Based on the survey data, FAA data, field data, and assumptions presented herein, the total estimated quantity of fuel discarded to the ground surface annu- ally from GA operations is approximately 125,000 gallons. Note that a majority of the input variables can be substantiated with data, observations, or sup- ported assumptions, except the reduction factor for touch-and-goes (Step 6 bullet point 2, currently 0.5). If low- and high-range reduction factors are applied for a sensitivity analysis (low = 0.3 and high = 0.7), the estimated range of fuel discarded to the ground annually by GA operations is 75,000 gallons to 175,000 gallons. It should be noted that the objective of the research was to estimate the discard amount that enters the stormwater runoff system from airports. It has long been argued or rationalized by pilots that discarded fuel will evaporate and thus not negatively impact the environment. The various fate and transport mechanisms for discarded fuel are discussed in the Discarded Fuel—Fate, Environ- mental Implications, and Applicable Regulations section. It should also be noted that the estimate provided in this Digest is the average quantity dis- carded to the ground surface, and not necessarily the quantity that enters the stormwater system. Deter- mining the quantity of discarded fuel that actu- ally enters the stormwater system would require extensive stormwater sampling, monitoring, and modeling, all of which are beyond the scope of this research project. Previous Estimates and Quantity in Context An estimate of 3 million gallons of aviation fuel discarded to the ground surface annually was found in the literature during the research (3). Although the objectives of the research project did not include eval- uating or checking previous estimates, the 3-million gallon estimate is quite inconsistent with the value identified herein, and was hence given a cursory

4review. The 3-million gallon value appears to be overestimated for the following reasons: • Overestimate of Sump Quantity—A per- flight sump quantity of 8.5 liquid ounces, or about 250 mL, was used. This is over eight times greater than the amount estimated in the current research (G = 30.1 mL). • No Reductions for Disposal Behavior—No reductions were incorporated for pilots who may take smaller samples, who may return the fuel to the fuel tanks, or who may not conduct a pre-flight sampling exercise, all of which were confirmed in the current research. • Not All Take-offs Involve Sampling—No reductions were incorporated for the fact that not all take-off operations involve a sampling event. • Not Plausible—Three million gallons is nearly 1% of the total aviation gasoline (avgas) con- sumed (377,000,000 gallons), as referenced in the calculation itself. As an example of implau- sibility, each average single engine GA aircraft refueling exercise (~50 gallons) would result in nearly one-half gallon discarded. This is not plausible and is inconsistent with the field research conducted as part of this research project. The estimated quantity of between 75,000 gal- lons and 175,000 gallons per year can be put into context with other petroleum discharge estimates such as the following: • 2.6 million gallons of oil is spilled per year, which impacts navigable U.S. waterways (4). • 16 million gallons of petroleum per year are released to the sea via stormwater runoff and rivers (5). • 17 million gallons of gasoline are spilled each year by overfilling lawn mowers (6). • 11 million gallons of oil were released by Exxon Valdez (7). • Over 200 million gallons of oil were released by the Deepwater Horizon incident (8). • 100 million gallons of oil are spilled each year in the United States (9). • The EPA asserts that 200,000,000 gallons of used oil are improperly disposed of annu- ally (10). To put the research quantity estimate into further context, consider that there are currently 5,295 pub- lic use landing facilities (i.e., airports) in the United States (J. Collins, email communication, December 31, 2013). If an assumption is made that the quan- tity of discarded fuel is equally divided among these public use landing facilities, this amounts to between 14 and 33 gallons discarded per airport per year. OBJECTIVE 1—SYNOPSIS Quantity Estimate The estimated range of fuel discarded to the ground annually by GA operations is 75,000 gallons to 175,000 gallons. This amounts to between 14 and 33 gallons dis- carded per airport per year. DISCARDED FUEL—FATE, ENVIRONMENTAL IMPLICATIONS, AND APPLICABLE REGULATIONS Sampled fuel that is discarded to the ground sur- face can interact and move in the environment in a number of ways. These interactions and movements are often called “fate and transport” mechanism. These mechanisms, along with key environmental impacts and applicable legal implications, are discussed in the following sections. For an appropriate fate and transport discus- sion, it is important to have a general understanding of avgas. Avgas can consist of over 150 chemicals, all of which behave differently in the environment once discarded. Avgas consists almost exclusively of compounds formed by hydrogen and carbon mol- ecules, and is hence considered a petroleum “hydro- carbon.” Avgas also consists of a minor amount of inorganics, most specifically lead, in the form of tetraethyl lead (TEL). The lead raises the octane, prevents engine knock, and prevents valve seating issues. The most commonly used form of avgas for GA, 100 octane low lead (100LL), can contain up to 2.12 grams of lead per gallon (11). Petroleum hydro- carbons in avgas behave quite differently depending on the complexity of their individual chemical struc- tures. The lighter hydrocarbons are more volatile (evaporative) while the heavier, oilier constituents are less volatile and more persistent in the terres- trial environment. The lead can also persist in the terrestrial environment.

5Primary Fate and Transport Mechanisms Once discarded to the ground surface, avgas can degrade, persist, or move in the environment in a number of ways. The primary fate and transport mechanisms are summarized below: Evaporation/Volatilization. This mechanism accounts for a majority of fuel loss upon surface dis- card; however, pilots should be aware that the dis- carded fuel does not simply disappear. Rather, the fuel is simply converted from the liquid phase to the vapor phase, where it can become an air pollutant. Volatilization rates are highly dependent on numer- ous factors, including temperature, pressure, atmo- spheric moisture, and fuel-specific characteristics. It is estimated that as much as 95% of petroleum can be lost to volatilization within hours of discharge (5). Volatilization as a method of dealing with discarded avgas is inconsistent with other similar industries (e.g., vehicle refueling stations, vehicle manufactur- ers) which are required by the 1990 Amendments to the Clean Air Act (CAA) to install various vapor recovery systems to control volatile organic com- pound (VOC) emissions from vehicle refueling and tank refilling operations. The dispensing facili- ties alone are estimated to spend over $88 million annually in the operation of Stage II vapor recov- ery systems (ESIS, Inc., unpublished white paper, October 2013). Degradation. The constituents of avgas that do not evaporate are left behind to persist in the envi- ronment. Based on numerous factors, this remain- ing mass may degrade by any number of physical, chemical, or biological processes. One such process which has been successfully employed for remedia- tion of petroleum hydro-carbon is biodegradation, or degradation mediated by microorganisms. Bio- degradation requires optimum environmental con- ditions for the complete degradation of petroleum hydrocarbons to harmless byproducts (e.g., CO2, H2O). These environmental conditions include the presence of the appropriate type of indigenous micro- organisms, temperature, moisture, oxygen, and nutrients. Biodegradation is a relatively slow pro- cess (e.g., months, years) and is not believed to be an adequate response to discarded fuel based on the rates and reliance on optimum conditions. It should be noted that, although lead is an inorganic and in- organics are generally not biologically degradable, some research has shown that TEL, which has an organic component, can be biologically degraded (12). Compounds that do not degrade are left to per- sist in the environment until a transport mechanism carries them away as discussed below. Infiltration. Infiltration occurs when avgas is trans- ported vertically downward, either by its own gravi- tational forces or by the driving force of precipitation. Infiltration is generally limited to permeable surfaces (e.g., grass tie-downs). However, aviation fuel can chemically soften the asphalt binder material of paved areas and degrade the asphalt itself, creating localized permeable areas at permanent tie-downs (13). This localized permeability can then lead to transport (infiltration) of petroleum hydrocarbons to the underlying soil and groundwater. Stormwater. Avgas residuals that are not evapo- rated, degraded, or transported to the subsurface remain available for transport via stormwater run- off. Transport can occur by dissolving the lighter fractions of the fuel within the stormwater or by physically transporting the heavier and particulate fractions with the stormwater flow. Depending on the stormwater management features of the airport, this hydrocarbon-containing stormwater can be tempo- rarily detained in stormwater management features (e.g., basins, treatment wetlands), can be carried to permeable areas for infiltration, or can be transported to adjacent streams, lakes, wetlands, or other similar receiving water bodies. Environmental Implications Petroleum hydrocarbons in stormwater runoff, even at low concentrations, are known for their acute toxicity to aquatic organisms in receiving streams (14). At even lower concentrations where acute toxicity might not be immediately evident, research has shown long-term impacts to aquatic organisms (15). Metals in urban stormwater have the potential to impact water supplies and cause acute or chronic toxic impacts for aquatic life (14). Despite evidence of ecosystem con- sequences resulting from stormwater runoff, identify- ing human health impacts from stormwater runoff are not yet possible (15). Applicable Laws and Regulations Laws and regulations that may govern GA fuel sampling and disposal practices can come from the

6federal, state, and/or local level and can be quite com- plex and subject to interpretation. The following sec- tions provide an overview of potentially applicable regulations and how they may impact GA fuel sam- pling practices. However, it should be noted that this Digest does not provide a full legal review and GA pilots are advised to independently review any and all applicable laws and regulations. Federal regulations. The most applicable federal laws and regulations regarding GA fuel-tank sample disposal are likely the CAA and the Clean Water Act (CWA), both of which are implemented by the U.S. Environmental Protection Agency (EPA), although both are more locally administered and enforced by the individual states. The CAA applies primarily to large industrial sources (e.g., power plants, chemi- cal manufacturers, refiners) and motor vehicles, although smaller pollution sources (e.g., gas sta- tions, paint shops) are also regulated. No direct reference was found in the CAA regarding air pol- lution restrictions from the evaporation of GA fuel samples, although as noted earlier, gas stations and vehicle manufacturers are required to install vapor recovery units to capture gasoline vapors from these sources. It is unlikely that the quantity of GA fuel sample discards is a significant regulatory vapor source, given that this discard quantity (125,000 gallons) is only 0.00009% of the 133 billion gal- lons of gasoline dispensed and consumed in the United States in 2012 (16). Regarding the CWA, the most significant regula- tion applicable to GA fuel sampling and disposal is likely the stormwater provisions established in the CWA’s National Pollution Discharge Elimination System (NPDES). The CWA made it unlawful to dis- charge any pollutant from a point source into navi- gable waters, unless a permit was obtained, and the 1987 amendments to the CWA broadened the EPA’s permit coverage to include stormwater discharges in addition to point sources. The NPDES program is administered through a series of discharge permits, monitoring requirements, and, in some cases, efflu- ent limitations. Although ground discharge of sam- pled GA fuel could be construed as a violation of the CWA as defined above (i.e., “. . . discharge . . . into navigable waters . . .”), sampling of stormwater runoff from airports would be required to determine if the stormwater contained pollutants that would exceed certain CWA thresholds. It should be noted that NPDES permits are grouped by industry sector, and NPDES permits are only required for airports with de-icing operations. Included in the federal regulations research was an inquiry to the local FAA flight standards district office (FSDO) in Philadelphia. The FSDO inspector indicated that there are no known FAA regulations or restrictions on the proper method for disposing of sampled GA fuel and recommended that the FAA Airplane Flying Handbook be consulted for addi- tional clarity. The Handbook (FAA-H-8083-3A) was reviewed and no detail on how to dispose of fuel sam- ples was included, although instruction on draining fuel during preflight and after refueling was covered. State regulations. Other than the authority to imple- ment the CAA and CWA as described above, states generally have their own set of environmental laws and regulations. State regulations regarding GA fuel sample disposal can be aviation-specific or could loosely fall under other regulated industry sectors such as aboveground storage tanks or general envi- ronmental protection. A state-by-state review of all applicable regulations is well beyond the scope of this research project; however, the following examples were identified during the research that indicate either a quantity-based direct ground discharge restriction or a general intolerance toward such practices: • Florida Department of Transportation (FDOT)—“Florida law prohibits dumping ‘sumped’ aviation fuel on the ground (soil, pavement, or waterway). Violators are subject to a fine of up to $50,000. See Section 403.727, Florida Statutes for details.” (17). • New York State Department of Environmen- tal Conservation (NYSDEC)—“Petroleum spills must be reported to DEC unless they meet all of the following criteria (18): – The spill is known to be less than 5 gallons; and – The spill is contained and under the control of the spiller; and – The spill has not and will not reach the State’s water or any land; and – The spill is cleaned up within 2 hours of discovery.” • Maryland Department of the Environment (MDE)—“A person discharging or permitting the discharge of oil, or who either actively or passively participates in the discharge or spill- ing of oil, either from a land based installation, including vehicles in transit, or from any vessel,

7ship or boat of any kind, shall report the incident immediately to the administration.” (Taylor GeoServices, Inc., unpublished work, October 2006). Note that no minimum quan- tity is specified, and this has been corroborated by communication with personnel familiar with gasoline station reporting requirements that MDE requires all releases, regardless of quantity, to be reported (J. Worth, email com- munication, November 7, 2013). • Other States. Several states have an identi- fied minimum reportable release quantity for petroleum hydrocarbons (e.g., gasoline, oil, etc.), below which reporting is not required, but under the assumption that the release does not impact navigable waters and does not cause a sheen or film on the water surface. Some exam- ples include Delaware, Ohio, and Virginia, all of which have a minimum reportable quantity of 25 gallons (Taylor GeoServices, Inc., unpub- lished work, October 2006; J. Worth, email communication, November 7, 2013). Local regulations. Local regulations can come in two forms: (1) municipal-specific ordinances and environmental regulations or (2) airport-specific environmental plans such as Stormwater Pollution Prevention Plans (SWPPP) or Spill Prevention, Control, and Countermeasures (SPCC) Plans. Some airport-specific plans contain guidelines or recommendations for proper fuel sample management. For example, Seattle-Tacoma International Airport’s SWPPP (19) provides fuel storage and delivery best management practices in Table 09 by stating “Use GATS jars to take fuel samples. Dispose of samples at designated collection sites. Use fire-rated con- tainers for storage of fuel samples.” This statement was contained near the end of the SWPPP, and it is unclear whether these types of documents have the force of law, and if so, who enforces them and how a GA pilot would know to review such a plan. CURRENT STATE OF THE PRACTICE The research conducted herein has shown that there is a wide range of fuel sampling and disposal practices currently being employed, and no clear industrywide consensus of how the sampling and disposal should be conducted. Although the ultimate goal of this research project is to develop fuel sam- pling best practices, it is important to first review the current state of the practice as a baseline to iden- tify potential improvements. The current state of the practice was identified via the following four dis- crete research methods, all of which are described in more detail in the following subsections: 1. Literature and Internet research. 2. Inquiries to various aviation groups. 3. Field-based observations. 4. Pilot survey. Literature research. Literature research included scouring and querying numerous scholarly databases (including the National Academies TRID Database), regulations, Internet sites, aviation manuals, pilot operating handbooks (POH) and videos. A majority of the information obtained from this research is incorporated into the text of this Digest as appropri- ate. However, as it applies specifically to the state of the practice, the following noteworthy items were identified: • Aircraft POHs—Numerous POHs were reviewed, and although all made reference to pre-flight fuel sampling procedures, all were silent on how to manage the fuel samples once collected. • Aviation Manuals—Pilot training and oper- ating manuals were reviewed, and as with the POHs, reference was made to sampling but not to sample disposal. • Online Videos—A cursory search of YouTube was conducted using the keyword “preflight.” Of the first five videos where GA pre-flight fuel sampling was presented, three of the five videos (60%) showed the ground discarding of the fuel samples (20). • Previous Work—The literature research revealed that the fuel sampling and disposal issue is not new and has been evaluated and debated for several years. For example, the quantity estimate of 3 million gallons previ- ously mentioned was originally calculated in 1989. Additionally, aviation associations such as AOPA have published articles on the topic as far back as 2003. Furthermore, a proposal was prepared by the Aviators Model Code of Conduct in 2006, for consideration by FAA’s Aeronautical Charting Office, to add “Collec- tion Sites” to all airport diagrams such that pilots knew where to properly dispose of fuel samples, used oil, and other similar wastes

8(10). Follow-up documentation from the FAA indicates that the issue was considered and declined and the issue closed (21). Inquiries. Phone, email, personal, and/or website inquiries were made to several aviation associa- tions, fixed-based operators (FBOs), flight schools, and airport managers to identify the fuel sampling and disposal procedures in practice at various fa- cilities. Results of the inquiries are summarized on Table 1 and show a wide range of best practices and responsiveness, but a clear effort on the part of many organizations to properly manage fuel samples. Field observations. Field-based research was con- ducted at three GA airports by positioning a field technician at a central location for 4-hour blocks of time (8 AM to Noon) on three consecutive Saturdays. The intent of the field research was to qualitatively ground-truth the sampling/disposal responses from the pilot survey (see below), and not necessarily to generate quantitative research data. As shown in Table 2, limited sampling/disposal operations were observed during the field research. However, for those operations observed, they confirmed a range of disposal habits, including return of fuel to the aircraft fuel tanks (37%) and ground discarding (63%). The Table 1 Summary of inquiries. Date Method of Inquiry Type Response? Response Notes/Findings 9/4/2013 Phone Airport YES Five flight schools on the field. Airport has provided disposal containers about the field for pilots, but they never get used. Airport sells ~300,000 gallons of avgas per year. 9/9/2013 email Aviation Association Yes Association-owned aircraft are equipped with GATS jars to return sumped/sampled fuel to the fuel tank. Each hangar is also equipped with a disposal can. This is a written policy to be included in the next revision of the policy manual. Association also publishes safety advisories for its constituents on how to best manage sumped/ sampled fuel. 9/9/2013 Phone and site visit FBO/Flight School/ Charter YES Facility provides 5-gallon buckets for fuel disposal. New pilots are taught to dispose in the bucket or return to the fuel tanks. Impetus for the disposal practice is to preserve the pavement, as advised by the pavement contractor. Fuel in the bucket always evaporates and doesn’t accumulate. 10/2/2013 Phone FBO YES FBO provides a fuel disposal container, but pilots don’t use it and it is always dry. Container was placed at the request of the airport authority. 10/8/2013 Phone Airport YES PI called based on a presentation found online by the airport environmental manager. One of the slides indicated a ranking of “[Airport’s] Highest Ranked Impacts” and #3 was fuel sumping by GA (3 of 7). PI asked if this was supported by data, and manager indicated no, it was a gut feel. Airport has disposal containers about the airport but they are not used. An action item on manager’s list is to remind the FBOs/flight schools of the containers. No metals, oil and grease (O&G), or VOCs are ever detected in their NPDES stormwater samples. The driver for the containers was good environmental stewardship.

9observations also revealed that few pilots sump their tanks following refueling. No observations of fuel vent leakage or tank overfilling were witnessed. Pilot survey. The electronic pilot survey provided data about overall pilot sampling/disposal behav- iors (in addition to the discard quantification data). Specifically, question 7 (What do you do with the fuel sample(s) once they have been collected and inspected?) revealed that close to 40% of pilots dis- card the fuel samples to the ground regardless of the sample condition, whereas approximately 60% of pilots manage the fuel samples by either returning them to the aircraft fuel tanks or disposing of them in a ground-based container. The field-based obser- vations corroborate the pilot survey data (i.e., pilots engage in a range of disposal behaviors); however, the percentages do not totally correlate. The field- based research showed approximately 63% of pilots discard the fuel samples to the surface, whereas the pilot survey data showed approximately 40% of pi- lots engage in this behavior. This data disparity is considered minor and is not unexpected, given the responses from the pilot survey and the inherent ten- dency of respondents to provide a positive response (e.g., I return my samples to the tank), despite the anonymity of the survey. OBJECTIVE 2—BEST PRACTICES Regardless of the estimated quantity of fuel dis- carded to the surface as determined in Objective 1, or the potential impact this practice may have on the environment or the airport infrastructure, this prac- tice is economically wasteful, legally questionable, and easily avoidable with the implementation of any number of best practices. The term best practices has many definitions, but can generally be defined as physical, structural, or managerial procedures that, when implemented, will achieve a common goal. Best practices are adaptive learning processes, as opposed to a fixed set of rules or guidelines that can be modified and improved with time and use. In the case of GA fuel sampling and disposal, the best practices can be separated into one of three categories: pilot-based, airport-based, and aviation community-based. The goals of these GA fuel sampling and disposal best practices are to mitigate environmental impacts, comply with applicable regulations and optimize fuel econom- ics. The following sections present several aviation best practices that can be used, either singly, or in combination, to achieve these goals. It should be noted that the recommendations for best practices provided in this Digest are in no way a confirmation that the current state of GA fuel disposal practices have, in fact, a documented impact on environmental media (e.g., surface water, groundwater, stormwater, soil, sediment, or air) or that implementation of these best practices will mit- igate any presumed impacts. A rigorous sampling and monitoring program would be necessary to con- firm any cause-and-effect relationship or to evalu- ate trends in stormwater quality improvement as a result of implementation of these best practices. Pilot Best Practices Pilot-based best practices are the most effective means of achieving the overall goals, given that the pilot is the source of the activity, and it is always most efficient to address issues at the source, as opposed to implementing subsequent measures downstream to counteract the initial behavior. Additionally, the regulations reviewed as part of this research gener- ally place the responsibility for discharges, clean up, and reporting on “. . . the person in charge . . .” or “. . . the person responsible for the discharge . . .” Pilot-based best practices include fuel sampling equipment that allows filtration and re-use of the fuel in the aircraft, and modifying pilot behavior to encourage the re-use, re-purposing, or proper dis- posal of fuel samples. These best practices are dis- cussed in detail below. Fuel sampling equipment. Field testing of three fuel sampling devices was conducted as part of the research project. A summary of the testing proce- dures and results is provided below. All three samplers employ essentially the same technology: a fine mesh screen which, when pre-wetted with fuel, permits the passage of uncontaminated fuel while holding back water or particulate. Fuel (100LL) was pre-dosed with water and tested in triplicate in each of the three sampling devices. Based on the testing, all three products were able to remove water and particulate contamination from the fuel samples while allowing the uncontaminated fuel to be returned directly to the aircraft fuel tanks. Additionally, all three devices were able to capture essentially all of the pre-dose water. As such, it can be presumed that this water (if encountered), albeit

10 Table 2 Summary of field observations. Date Time Annual Operations (airnav.com) and Tower/Non-Tower OPERATION TYPE AIRCRAFT TYPE FUEL SAMPLING/STRAINING OBSERVATIONS NOTES (include overflow and vent release observations)Refuel Pre-flight Other High/ Low Wing Single Twin Sumping Occur? (Y/N) Qty Sumps Sampling device Qty to Tank Qty to Ground Qty to Canister Cockpit fuel drain 8/31/2013 0840 24,820 - Non-towered X low X Y 4 test tube 1 3 0 0 Personal (Archer/Dakota) parked outside on ramp 8/31/2013 0900 24,820 - Non-towered X high X Y 3 unknown 0 2 0 0 Flight school 8/31/2013 1030 24,820 - Non-towered X low X N unknown N/A 0 0 0 0 Personal (Dakota) refuel at pumps 8/31/2013 1120 24,820 - Non-towered X high X N unknown N/A 0 0 0 0 Personal (Aeronca Champ) refuel at pumps 9/7/2013 0845 90,155 - Towered X high X N unknown N/A 0 0 0 0 Flight school. Flaps down, so not sure if plane was already preflighted and we missed sumping observation. Not included in totals. 9/7/2013 1130 90,155 - Towered X high X Y 2 test tube 2 0 0 0 Private overnight transient 9/14/2013 1015 27,010 - Non-towered X low X N unknown N/A 0 0 0 0 Experimental refuel at pumps TOTALS 8 N/A 3/8 (37%) 5/8 (63%) 0 0 likely containing some dissolved constituents of petroleum hydrocarbons, is mostly water and can therefore be discarded to the ground. Based on these results, these products are appropriate as a best practice for proper re-use of fuel. It should be noted that no regulations were encountered during this research project that indicate a restriction of fuel re-use in the aircraft. Pilot practices. Pilot best practices include re- using or properly disposing of the fuel samples. More details on each of these best practices are provided below: • Re-use fuel samples—The purpose of fuel sampling is to visually inspect the fuel for contamination. If visual contamination is not present, the samples can be returned to the aircraft tank with minimal risk of unforeseen micro-contamination. Note that over 92% of the pilot survey respondents indicated that they either never or very infrequently expe- rienced contamination in their fuel samples. This practice is distinctly separate from the use of filtration sampling devices as described in Pilot Best Practices subsection. • Dispose of fuel samples—If re-use is not an option, the fuel samples can be placed in an approved ground-based fuel container for future re-purposing or proper disposal. This best practice is discussed further in the Airport Best Practices section. • Re-purposing fuel samples—A review of regulations and inquiries to the EPA regard- ing any restriction of re-use, considering that 100LL contains low levels of lead, was per- formed. The CAA regulates fuel usage and contains restrictive language and civil penal- ties for leaded fuel usage in motor vehicles labeled with “unleaded gasoline only” (CAA Section 211(g)(1) Misfueling). Although Sec- tion 216(2) of the CAA defines motor vehicles as “. . . any self-propelled vehicle designed for transporting persons or property on a street or highway,” further review of background CAA documents and discussions with EPA personnel indicate that the topic of misfuel- ing does apply to non-road vehicles as well.

11 Specifically, the EPA provided proposed and final rule language (22, 23) from a recently- promulgated misfueling regulation pertain- ing to ethanol-containing fuel (E15). The U.S. EPA opined that the use of lead in non- approved engines would be considered simi- lar to the E15 scenario, including penalties for misfueling. Following these discussions with the EPA, it appears that re-purposing of leaded avgas in non-road, airport-based equipment could only occur under one of three conditions: (1) the equipment would have to be approved for leaded fuel use (i.e., an old engine manufactured prior to leaded fuel restrictions), (2) the user would have to demonstrate to the EPA, likely through rig- orous testing, that the equipment could still meet applicable emission regulations while using the leaded fuel, or (3) the user would have to make a legal argument that the CAA regulations do not apply to this condition. Assuming that the misfueling of E15 is akin to re-purposing of leaded fuel as discussed herein, the E15 proposed and final rules con- firm that misfueling violations can result in a civil penalty up to $37,500 for every day of each such violation. Airport Best Practices Various airport-based best practices can be implemented to achieve the quality goals. The prac- tices are interrelated and interdependent but can gen- erally be described as either physical, operational, or administrative. These are described below. • Physical—Physical best practices can include placement and usage of containers to tem- porarily store fuel samples until the fuel can either be properly disposed of or re-purposed. Containers should be Type 1 metal safety cans with self-closing spring-loaded caps and anti-flashback devices approved by the Occupational Health and Safety Administra- tion (OSHA) (1910.106(a)(29)). Individual can capacities cannot exceed 5 gallons, and no more than 25 gallons of gasoline can be stored collectively in safety cans outside of an Table 2 Summary of field observations. Date Time Annual Operations (airnav.com) and Tower/Non-Tower OPERATION TYPE AIRCRAFT TYPE FUEL SAMPLING/STRAINING OBSERVATIONS NOTES (include overflow and vent release observations)Refuel Pre-flight Other High/ Low Wing Single Twin Sumping Occur? (Y/N) Qty Sumps Sampling device Qty to Tank Qty to Ground Qty to Canister Cockpit fuel drain 8/31/2013 0840 24,820 - Non-towered X low X Y 4 test tube 1 3 0 0 Personal (Archer/Dakota) parked outside on ramp 8/31/2013 0900 24,820 - Non-towered X high X Y 3 unknown 0 2 0 0 Flight school 8/31/2013 1030 24,820 - Non-towered X low X N unknown N/A 0 0 0 0 Personal (Dakota) refuel at pumps 8/31/2013 1120 24,820 - Non-towered X high X N unknown N/A 0 0 0 0 Personal (Aeronca Champ) refuel at pumps 9/7/2013 0845 90,155 - Towered X high X N unknown N/A 0 0 0 0 Flight school. Flaps down, so not sure if plane was already preflighted and we missed sumping observation. Not included in totals. 9/7/2013 1130 90,155 - Towered X high X Y 2 test tube 2 0 0 0 Private overnight transient 9/14/2013 1015 27,010 - Non-towered X low X N unknown N/A 0 0 0 0 Experimental refuel at pumps TOTALS 8 N/A 3/8 (37%) 5/8 (63%) 0 0

12 approved storage cabinet, per OSHA. Other physical best practices may include storm- water management features constructed as part of an airport’s stormwater management and control infrastructure. Table 3 provides a summary of example stormwater control measures (SCMs) that can be used to con- trol, contain, and improve stormwater quality; however, it should be explicitly understood that the presentation of these SCMs in this Digest is in no way a suggestion that airports are required to or should install these features. Rather, if an airport is in the process of expan- sion or development, and SCMs are required as part of the development process, Table 3 pro- vides examples of SCMs which are consistent with stormwater runoff containing petroleum hydrocarbons. • Operational—Airport operational best prac- tices include those standard operating proce- dures that are either voluntarily implemented or required by regulation or permit to control or monitor environmental conditions. One such operational procedure is to advise airport re - fuel ing personnel that topping off or overfilling of aircraft tanks is discouraged. Topping off or overfilling of tanks can cause an immedi- ate release of fuel to the ground. Additionally, filling tanks to the very top, followed by an increase in ambient temperature, can cause the fuel to expand and release through the overfill vents. Avgas expands approximately 1% per 10 degrees Celsius (°C) of tempera- ture increase (24). Therefore, if a GA aircraft fuel tank containing 25 gallons is topped off at 10°C (50° Fahrenheit [°F]) and the tempera- ture increases to 30°C (86°F), one-half gallon of fuel can be released through the overflow system. This is not an insignificant quantity of fuel, but it is also a fairly extreme increase in ambient temperature (36°F) that is not expe- rienced frequently. Regardless, this overflow quantity underscores the need for refueling personnel and pilots to be aware of the envi- ronmental, regulatory, and economic conse- quences of filling aircraft tanks to the very top. • Administrative—Administrative best prac- tices involve the implementation of airport- based policies which are driven by the airport’s permits or other regulatory obligations. These policies are documented in manuals such as an SPCC plan or an SWPPP. Generally speaking, these types of administrative best practices are only as strong as the system of procedures and personnel charged with overseeing and enforc- ing them. This was evident during the research inquiries (see Table 1), whereby several facili- ties (airports) made attempts to encourage best practices (e.g., providing disposal cans) but there was a general acknowledgment that, in reality, the cans go unused. Other administra- tive best practices can include airport signage or postings to describe the appropriate method of fuel sampling and disposal and the negative impacts (or regulatory infractions) of ground discarding. Aviation Community Best Practices In the context of fuel sampling and disposal best practices, the aviation community is a broadly defined term that includes various stakeholders with a poten- tial interest in this issue, including flight schools, flight training organizations, aviation associations, and aircraft manufacturers. The best practices for this segment of the aviation community are gener- ally more onerous and far-reaching and may not be easily implementable without significant effort, cor- porate will, or external motivation. Regardless, this Digest would not be complete without a review of all potential best practices that can achieve the intended goals. Examples of aviation community best prac- tices include the following: • Flight Training Organizations—Primary flight training is the first opportunity to instill certain behaviors in student pilots. Through standardized fuel sampling best practices, the flight instructors can make a significant impact on future pilots. Therefore, flight instructors, flight schools, and especially the organizations that prepare the primary flight training manuals should modify their procedures to incorporate the fuel sampling best practices contained in this Digest. It should be noted that a trend in more appropriate fuel sample disposal practices (e.g., re-use or disposal as opposed to ground dis- card) was discernable in the pilot survey data when comparing the disposal behavior results (question 7) to the pilot’s primary training era (question 10). This is an indication that appropriate best practices are beginning to be

13 Table 3 Summary of example stormwater control measures. Stormwater Control Measures (SCMs) Description Treatment Effectiveness Capital Cost O&M Requirements Detention Basin Dry basin with outlet structure to control stormwater peak flow. No mechanism for water quality treatment. LOW LOW LOW Infiltration Basins, Beds, and Trenches Primarily for stormwater volume reduction via infiltration. Water quality treatment only occurs as an unintended consequence of infiltrating through underlying soil. Can result in groundwater contamination if sufficient treatment doesn’t occur. LOW HIGH LOW Wet Pond/Retention Basin Basin with permanent pool of water to support vegetation and promote treatment through sediment settling, evaporation, and transpiration. MODERATE LOW MODERATE Rain Garden/ Bioretention Engineered “depression” or basin with a subsurface layer of organics to support vegetation and promote filtration and biodegradation. HIGH MODERATE MODERATE Constructed Wetland Large treatment basin with permanent pool, vegetation, and varying “engineered” zones to provide treatment via physical and biological mechanisms. Generally requires large land area to construct. HIGH MODERATE MODERATE Engineered Filters and Structures Filters and/or hydrodynamic units in engineered basins or structures. Generally small units designed for small drainage areas. Includes proprietary pre-packaged treatment units. HIGH HIGH HIGH Oil-Water Separator Treatment unit with multiple chambers and weirs to separate and capture oil (free-phase, non- dissolved) in stormwater runoff. HIGH HIGH HIGH Note: all ratings (low/moderate/high) are for relative comparison purposes only regarding treatment of stormwater containing petroleum hydrocarbons. O&M: Operations and Maintenance.

14 implemented as part of primary flight training and this movement should be capitalized on and re-enforced moving forward. • Aviation Associations—Associations such as AOPA, Civil Air Patrol, and numerous others should disseminate (or continue to dis- seminate) this type of information to educate the aviation community and their individual constituents on the appropriate best practices. This can be accomplished via newsletters, email blasts, and/or webinars, to name a few. • Aircraft Manufacturers—These are undoubt- edly the most onerous of the fuel sampling best practices contained in this Digest, but as stakeholders, the aircraft manufacturers have a vested interest in appropriate fuel sampling and disposal practices. Best practices to be considered by aircraft manufacturers include the following: – Modify future POHs to address the appro- priate methods of managing pre-flight fuel samples. – Modify future aircraft designs to do away with remote or cowl-based fuel strainers that make fuel sample collection difficult. – Ensure that all future high wing aircraft have an easy means of accessing the fuel tank filler area to return fuel samples to the tank. – Modify future aircraft to have an integral fuel screen mechanism or provide an appro- priate filtering fuel sampler with each new aircraft. – Modify future aircraft to have fewer fuel tank sumps so there are fewer fuel sam- ples generated. Data from the pilot survey (question 2) indicate that some single- engine aircraft have up to 13 sump locations. CONCLUSIONS AND SUGGESTIONS Conclusions that can be drawn from this research project include the following: 1. An estimated range of between 75,000 gal- lons and 175,000 gallons of avgas is dis- carded to the ground surface each year by GA pilots. 2. Environmental impacts to air, surface water, groundwater, and soil occur from the release of petroleum hydrocarbons to the environ- ment. Environmental impacts from GA fuel discarding cannot be confirmed, quantified, or evaluated without further research. 3. Numerous local, state, and federal environ- mental regulations potentially apply to GA fuel sampling and disposal activities. A review of these regulations indicates that ground dis- carding of avgas may be in violation of certain regulations. 4. The current state of the practice for GA fuel sampling and disposal varies and includes a segment of the pilot community that discards samples to the ground surface (40% to 63%) while others re-use or properly dispose of the fuel samples. Trends in the survey data indi- cate a marginal temporal movement away from ground discarding for more recently trained pilots. 5. The value of the discarded avgas is between $450,000 and $1,050,000 annually (at $6.00/ gallon). OBJECTIVE 2—SYNOPSIS Best Practice Suggestions Pilot Best Practices (implement at least one) • Use filtering fuel samplers to return samples to the fuel tank • Re-use fuel samples in the aircraft if not contaminated • Properly dispose of fuel samples Airport Best Practices (implement any that are applicable and practical) • Provide fuel disposal containers • Avoid overfilling and topping off of aircraft tanks • Implement, encourage, and enforce proper disposal practices Aviation Community Best Practices (consider implementing) • Teach proper disposal practices at the primary flight training level • Disseminate proper disposal practices via avi- ation associations • Consider airframe, engine, and pilot operating handbook (POH) modifications in future air- craft designs to encourage proper fuel disposal

15 REFERENCES 1. SurveyMonkey® (2013). www.surveymonkey.com/ s/GA_Fuel_Sampling_Survey. 2. Federal Aviation Administration (2013). www.faa. gov/airports/airport_safety/airportdata_5010/menu/ index.cfm. 3. 1989 Detrimental Impact Study of Aircraft Fuel Sam- pling and Year 2000 Followup Addendum (2000). http://www.reidhillview.com/Lead_4_times_car_ fuel.htm. 4. U.S. Department of Transportation (2012). Table 4-54: Petroleum Oil Spills Impacting Navigable U.S. Water- ways. Research and Innovative Technology Statistics, Bureau of Transportation Statistics, Washington D.C. 5. The National Research Council of the National Academies. Oil in the Sea III, 3rd ed. The National Academies Press, Washington, D.C., 2003. 6. The Christian Science Monitor (2008). Spilled lawn mower gas costs you and the environment. http:// www.csmonitor.com/The-Culture/Gardening/diggin- it/2008/0827/spilled-lawn-mower-gas-costs-you- and-the-environment. 7. U.S. Department of the Interior, National Park Ser- vice (2009). 20 Years Later . . . Exxon Valdez Oil Spill. Kenai Fjords National Park, Seward, AK, March 1. http://www.nps.gov/kefj/naturescience/ upload/KEFJ_EVOS_1989-2009_qa.pdf. 8. U.S. Environmental Protection Agency, National Response Team (2011). On Scene Coordinator Report, Deepwater Horizon Oil Spill. September. 9. AbsorbentsOnline.com (2013). http://www.absor bentsonline.com/oilspillbasics.htm. 10. Aviators Model Code of Conduct and Michael Baum (2006). Proposal to Designate Locations of Collection Sites—Aeronautical Charting Forum. http://www. secureav.com/ACF.pdf. 11. Aircraft Owners and Pilots Association (2010). Issues related to Lead in Avgas. http://www.aopa.org/Media- Relations/Position-Papers/Issues-related-to-Lead-in- Avgas. 12. Teeling, H., and Cypionka, H. Microbial degradation of tetraethyl lead in soil monitored by microcalorim- etry. Applied Microbial Biotechnology, Vol. 48, 1997, pp. 275–279. 13. Hajek, J., Hall, J.W., and Hein, D.K. ACRP Synthesis of Airport Practice 22: Common Airport Pavement Main- tenance Practices. Transportation Research Board of the National Academies, Washington, D.C., 2011. 14. U.S. Environmental Protection Agency (1999). Pre- liminary Data Summary of Urban Storm Water Best Management Practices. Office of Water, Washing- ton, D.C., August. 15. Committee on Reducing Stormwater Discharge Contributions to Water Pollution, National Research Council Urban Stormwater Management in the United 6. The localized degradation of airport infra- structure, namely asphaltic pavement, occurs as a result of the discard activity. Suggestions developed as part of this research include implementation of various best practices and potential future research on the topic as follows: 1. Pilots should implement at least one of the following best practices: a. Use filtering fuel samplers to return filtered fuel samples to the aircraft. b. Re-use fuel samples in the aircraft if not visually contaminated. c. Properly dispose of fuel samples that are either contaminated or are otherwise not amenable to re-use or re-purposing. 2. Airports should consider implementing any of the following best practices which may be appli- cable to the facility and practical to implement: a. Place fuel disposal cans at locations about the airport that are convenient for and will be used by pilots for fuel sample disposal. b. Instruct aircraft re-fueling personnel not to top off or overfill aircraft fuel tanks. c. Implement and enforce best management practices or other procedures that are included in the airport’s SWPPP and/or SPCC plan. d. Prepare and post signage or other educa- tional materials at key airport locations informing the aviation community about fuel sampling and disposal best practices and the potential implications of ground discarding. 3. Aviation community stakeholders should con- sider implementing some of the best practices presented in this Digest for the stability and betterment of the GA sector. 4. Suggestions for future research or best prac- tices support include the following: a. Fund additional research, specifically air- port stormwater sampling and monitoring, to identify if the current fuel sample dis- card practice is, in fact, having an appre- ciable impact on the environment. b. Fund the preparation of generic, informa- tive signage and educational materials on the topic of fuel sampling best practices that can be distributed to and used by GA airports. c. Disseminate the information in this Digest to the aviation community.

16 21. Federal Aviation Administration (2006). Government/ Industry Aeronautical Charting Forum 06-01. FAA Control Number 06-01-180, April 19–20. 22. U.S. Environmental Protection Agency (2010). Regula- tion to Mitigate the Misfueling of Vehicles and Engines with Gasoline Containing Greater Than Ten Volume Percent Ethanol and Modifications to the Reformulated and Conventional Gasoline Programs. Federal Regis- ter, Proposed Rule, Vol. 75, No. 213, November 4. 23. U.S. Environmental Protection Agency (2011). Reg- ulation to Mitigate the Misfueling of Vehicles and Engines with Gasoline Containing Greater Than Ten Volume Percent Ethanol and Modifications to the Reformulated and Conventional Gasoline Programs. Federal Register, Vol. 76, No. 142, July 25. 24. MacDonald, S., and Peppler, I. Chapter 10– Airmanship. In From the Ground Up (Millennium ed.). Aviation Publishers Co., Ottawa, Ontario, Canada. 2004, pp. 265. States. The National Academies Press, Washington, D.C., 2009. 16. U.S. Department of Energy, Energy Information Ad- ministration (2013). FAQ–How much gasoline does the United States consume? Washington, D.C. http:// www.eia.gov/tools/faqs/faq.cfm?id=23&t=10. 17. Florida Department of Transportation, Aviation and Spaceports Office (2013). Aviation Environment. http://www.dot.state.fl.us/aviation/environment.shtm. 18. New York State Department of Environmental Con- servation (accessed 2013). Technical Field Guidance: Spill Reporting and Initial Notification Requirements. www.dec.ny.gov/docs/remediation_hudson_pdf/1x1. pdf. 19. Port of Seattle (2011). Seattle-Tacoma International Airport Stormwater Pollution Prevention Plan. August. 20. YouTube, LLC (2013). www.youtube.com/results? search_query=preflight&sm=3.

Subscriber Categories: Aviation • Environment These digests are issued in order to increase awareness of research results emanating from projects in the Cooperative Research Programs (CRP). Persons wanting to pursue the project subject matter in greater depth should contact the CRP Staff, Transportation Research Board of the National Academies, 500 Fifth Street, NW, Washington, DC 20001. COPYRIGHT INFORMATION Authors herein are responsible for the authenticity of their materials and for obtaining written permissions from publishers or persons who own the copyright to any previously published or copyrighted material used herein. Cooperative Research Programs (CRP) grants permission to reproduce material in this publication for classroom and not-for-profit purposes. Permission is given with the understanding that none of the material will be used to imply TRB, AASHTO, FAA, FHWA, FRA, FTA, or Transit Development Corporation endorsement of a particular product, method, or practice. It is expected that those reproducing the material in this document for educational and not-for-profit uses will give appropriate acknowledgment of the source of any reprinted or reproduced material. For other uses of the material, request permission from CRP. ISBN 978-0-309-28413-4 9 780309 284134 9 0 0 0 0 Transportation Research Board 500 Fifth Street, NW Washington, DC 20001