Wake Turbulence: An Obstacle to Increased Air Traffic Capacity (2008)

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.

3 Technical Challenges in Wake Turbulence Research Introduction There are two approaches to reducing the adverse impacts of wake turbulence on airport capacity while maintaining or enhancing the cur- rent level of safety. The ideal solution to the wake turbulence problem is to remove the threat by removing the wake. Although elimination of the vortex is physically impossible, it does appear possible to alleviate some of its effects. This requires innovative concepts that can be implemented on a significant number of aircraft. However, the benefits of vortex alle- viation can only be realized in the long term, and even then, only if a particular concept proves viable. If the threat cannot be eliminated, an improved spacing system can enable the next-best solution: Avoid the vortex or ensure that it can be tolerated. This requires gathering information about the position and strength of the vortex through some combination of monitoring and mod- eling. Using these data, separations can be safely reduced whenever the vortices have been carried away from the flight path or undergone suf- ficient breakup/decay. This solution will also require information about the local weather and its projected changes. The benefits of this approach are limited to specific atmospheric conditions, but can be realized imme- diately as they are implemented on an airport-by-airport basis without having to alter the aircraft fleet. A further enhancement to this type of solution is to provide the pilot with the information necessary to safely avoid nearby vortices. This could require altering the aircraft fleet to 29 

30 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY provide sensors, models, data display, local weather information, and information about the paths of preceding aircraft. Advances to spacing systems and vortex alleviation will both require specialized research and technical advances in vortex modeling, vortex measurement, meteorological measurement, and vortex visualization. Supporting studies such as hazard boundary definition, system-level benefit studies, and the development of a system to gather data about wake events will be key to targeting research and eventually implement- ing solutions. Not only do they require the same supporting research, but alleviation and spacing also are not mutually exclusive—the best possible system may include elements of each. This section presents these technical challenges in more detail, along with relevant findings and recommendations. First, however, a caveat: Because this report focuses on the wake vor- tex research needed for enabling capacity increases, approach and landing issues and, to a lesser extent, takeoff wake vortex issues were the main considerations. However, en route spacing reductions may also prove necessary to accommodate increased demand. The spectrum of aircraft to be commonly used in high-altitude jet flight will grow as the A380, at 1.25 million pounds maximum takeoff weight, is introduced at the high end, and “very light jets,” at perhaps as low as 6,000 pounds takeoff weight, are introduced at the low end. New safety challenges may also arise with the 1,000-ft minimum vertical spacing procedures recently adopted in many parts of the world for IFR operations. The FAA’s Web site contains more details: <www.faa.gov/about/office_org/headquarters_offices/ ato/service_units/enroute/rvsm/>. The JPDO should not ignore these issues. Fortunately, most of the research challenges presented here may also be applicable to en route spacing. Finding 3-1. En route wake vortex issues may arise, especially for very light jets in cruise. Recommendation 3-1. The JPDO should investigate and define specific requirements for research on the impact of cruise-altitude-generated wakes on capacity (including climb and descent) to avoid future prob- lems as fleet diversity increases. IMPROVED Spacing System Design Wake vortex considerations affect aircraft spacing standards for en route, approach and landing, arrival and departure. The FAA Research, Engineering and Development Advisory Committee’s (REDAC’s) Separa- tion Standards Working Group (SSWG) found that “the current system, 

TECHNICAL CHALLENGES 31 based on the separation standards that have evolved over the last 50 years, is safe, but still unable to meet projected demand. The separation standards (and the approach to establishing separation standards) now need to be reconsidered in order to meet the demand for increased capac- ity.” Furthermore, it also found that “in considering the possibilities for reducing separation standards, wake turbulence becomes the driving consideration. For NGATS [NextGen], wake turbulence could become the primary limiter of capacity” (REDAC SSWG, 2006). A four-dimensional, trajectory-based air traffic separation manage- ment system for arrival, departure, and en route conditions is a key com- ponent of the JPDO NextGen vision. This system would be based on pre- cise trajectory and position information, including short-term intent data, and performance-based separation standards that account for aircraft capabilities, the local environment (e.g., wake vortex), and the encounter geometry. This system would include a substantial degree of automation (JPDO, 2007). While such a system is clearly critical to achieving NextGen capacity goals, it is not clear that any coordinated program exists to tackle the necessary research, particularly on wake turbulence. Finding 3-2. NextGen is expected to include separation management capable of permitting reduced and dynamic separation standards. Recommendation 3-2. JPDO should conduct a detailed analysis of what wake turbulence research and development is needed to achieve its sep- aration management capability goals, and provide a detailed plan with milestones that will lead to successful development in the required time frame. Air traffic spacing system design, combined with progress in other technical challenges such as hazard boundary definition, wake modeling, wake visualization, and wake alleviation (both active and passive), has the potential to offer capacity increases in the short term and in the long term. Some of these solutions may be dynamicthat is, spacing requirements are situationally dependentbut static solutionsthat is, rule changes such as recategorization of aircraft classifications for spacingcan also provide benefit. A simple example of a situationally dependent system is to allow spacing to be reduced when the vortices are being transported out of the next aircraft’s path by the wind. Concepts that use crosswinds to enable closely spaced parallel approaches (CSPAs) and reduced separations on single-runway departures have been developed and tested at STL and IAH, respectively. Their operational implementation awaits approval of the safety cases, which is expected to occur soon (Lang, 2007). Currently, 

32 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY the FAA and NASA’s Airspace System Program are conducting work on elements of these simple dynamic spacing system designs. Closely spaced parallel departures and single-runway arrivals may be considered as well, although a MITRE-NASA study found that procedure changes to CSPAs are generally more promising owing to their less-stringent crosswind requirements (Lunsford et al., 2005). In the medium term, the focus is on ATC data-driven procedural changes, such as static aircraft wake class recategorization. In the long term, the goal is to develop more complex dynamic spacing that depends on weather and/or traffic mix. The ulti- mate vision is to achieve active wake avoidance solutions, including a dynamic wake separation/dynamic wake class categorization. However, this ultimate vision is one of the topics that lacks sufficient investment to achieve the NextGen goals. NASA’s Aircraft Vortex Spacing System (AVOSS) project, conducted at Dallas/Fort Worth International Airport (DFW) in the late 1990s, pro- vides a case study of the benefit that might be derived from reduced spacing, assuming that the airport is able to handle the additional capac- ity. This system used weather data to predict the behavior of a wake and the appropriate spacing for a following aircraft. The location of the wake was also measured with lidar. AVOSS was intended for research observa- tions; it had no capability to affect aircraft spacing. However, compari- sons between AVOSS’s guidance and actual throughput showed that an operational version of AVOSS could have increased capacity by up to 16 percent, with an average of about 6 percent. Reduced spacing offers the most benefit to capacity-constrained airports; O’Connor and Rutishauser (2001) estimated that if DFW were at 90 percent capacity, a 6 percent increase in capacity would result in a 40 percent decrease in delays. In addition, since delays propagate through the air transportation system, a decrease in delays at large, capacity-constrained airports would echo across the system. One of the major limits to the capacity enabled by such a system is its weather-dependency. Any dynamic wake vortex spacing criteria must be applicable a very high percentage of the time. Without this attribute, capacity gains will be offset by the inability to maintain schedule integrity when it is necessary to revert to today’s wake vortex spacing separation standards. Recategorization of some aircraftchanges to the weight boundaries for large, medium, and small classes, and perhaps also changes to the number of classes and the factor or factors that establish themcould reduce separation requirements in some operating circumstances. The development of procedures applicable to these specific types of opera- tions may allow some capacity gains in the medium term. R&D has pro- duced some capability to detect and track wake vortices in very localized 

TECHNICAL CHALLENGES 33 areas. Achieving more substantial long-term capacity gains will require more effective wake vortex prediction, detection, and display capabilities, both on the ground and in the aircraft. In the long term, aircraft modifica- tion to alleviate the wake vortex threat may begin to generate even further capacity. The European Commission has been pursuing very similar goals using a long-term, methodical procedure. It has already completed p ­ rojects focused on encounter modeling (WAKENC), safety assessment (S-WAKE), and operational implementation (ATC-WAKE). The FAA and NASA are closely involved in the commission’s current project, CREDOS, which examines the operational feasibility of wind-dependent separa- tions for departure operations. Data from the STL tests are used for this project; the FAA has also made one of its lidar systems available to collect data at Frankfurt International Airport (EDDF). This is a strong, mutually beneficial partnership that will lead, it is hoped, to an International Civil Aviation Organization rule change that takes advantage of this work (Erikson, 2007). Finding 3-3. Reducing the spacing needed to avoid wake turbulence will allow more efficient usage of existing runways. Recommendation 3-3. The FAA, assisted by NASA, should continue its current improved spacing programs, which promise results in the short and medium terms. In addition to allowing increased usage of existing runways in IMC conditions, dynamic spacing for CSPAs may eventually support a reduc- tion in the current wake-vortex-related runway separation standards of 3,400 ft for independent approaches (spacing needs to be maintained only between aircraft on the same approach path) or 2,500 ft for depen- dent approaches (spacing must be maintained between aircraft on the same approach path and aircraft on adjacent approach paths). Reducing the required spacing of CSPAs will, in turn, allow construction of new runways between existing ones. Building new runways within existing airport boundaries will have less environmental impact than will expan- sion, resulting in a shorter, less expensive approval process. For airports whose borders may be constrained by geographical features (such as water or mountains) or heavy development, this is the only alternative to adding runways. The FAA should continue these sorts of development efforts so that the minimum runway spacing can be determined for future airport expansions and the required suite of technology to support the reduced spacing can be established. These efforts will require system-level studies 

34 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY TABLE 3-1  Milestones for Advanced Spacing System Design Time Horizon Milestone Short term Complete tests of conditional spacing reduction to closely spaced parallel approach (CSPA) at individual airports Medium term Allow conditional spacing reduction to CSPA at all selected airports Begin tests of conditional spacing reduction for single runway approaches Integrate reliable weather sensors into CSPA Long term Allow conditional spacing reduction to CSPA at all airports Allow conditional spacing reduction for single runway approaches at selected airports Begin tests of GPS-enabled, en route dynamic spacing as well as simulator and flight experiments to gain acceptance by pilots, controllers, operators, and the FAA. Because it can take 10 years or longer to plan and design a new runway (GAO, 2003), it is imperative that this work start early in order for airport managers to understand their runway spacing requirements. Benefit studies performed in the short term can identify runway construction strategies capable of producing significant long-term gains in capacity. Paradoxically, such studies may allow identi- fying situations where runways should be aligned at an angle (instead of parallel) to the dominant wind direction, so that wakes can be more often transported away, increasing the percentage of time during which closer spacing could be permitted. Milestones for advanced spacing system design are shown in Table 3-1. Finding 3-4. In many airports, the use of dynamic wake vortex spacing standards could permit parallel runways to be built closer together, meaning that in some cases, new runways may be built between exist- ing runways, or elsewhere on existing airport property. Recommendation 3-4. The FAA should pursue work in the short and medium terms to determine minimum runway spacing for future air- port expansion, which is needed to achieve capacity goals in the long term. vortex visualization: cockpit and controller Dynamic spacing of aircraft based on wake vortex motion will require prediction of wind behavior over roughly the next hour. It also will be necessary for the pilot and/or the controller to have information on the 

TECHNICAL CHALLENGES 35 wake position in real time as a safety net to verify the predicted separation provided. The information could be presented to the pilot numerically, visually, or by a simple red-light/green-light system. One option for pre- senting the real-time wake position that has been researched is visualiza- tion. A visualization capability will give pilots and controllers better situ- ational awareness, improving safety and allowing closer spacing. Ideally, real-time information on the position of the wake would be obtained by direct measurement. Near the ground, this may be possible to measure the position of the wake; however, it would be contingent on the develop- ment of all-weather sensors. Airborne sensors for measuring the wake of neighboring traffic are similarly unavailable. It is more likely that a real- time prediction model of the wake could be developed based on real-time traffic position broadcasts, aircraft characteristics, and measurements of environmental parameters such as wind and turbulence, supported by measurement of atmospheric and meteorological parameters. Real-time prediction models yield large variations in the position of the wake behind an aircraft, although most of the variation is strongly dependent on parameters that potentially could be determined by sen- sors or other means. With sufficient modeling development along with the associated error bounds, this knowledge can be used to present the predicted wake hazard zone to pilots and controllers. As discussed in the section “Wake Vortex Modeling,” real-time m ­ odels have been developed and are continually being refined. Hahn et al. (2004) applied real-time models for prediction of the wake danger zone, and Holforty (2003) compiled a model that predicted the wake loca- tion for a visualization flight test experiment. In addition, Ma and Zheng (1994) showed a real-time model and a method of presenting a three- dimensional visualization. A real-time model requires several param- eters: the three-dimensional position of the wake-generating aircraft, air mass motion, and the characteristics of the wake-generating aircraft. The accuracy of the wake location prediction will depend primarily on the number and accuracy of the above parameters that are available for input- ting to the real-time model. This model can determine the position in three dimensions of the wake danger zone for depiction by neighboring aircraft or air traffic controllers. The danger zone would be determined by projecting the hazard boundary onto the nominal wake location and accounting for any uncertainty. The size or existence of the danger zone would also depend on the relative size of the two aircraft. For example, a small, single-engine aircraft would not create a danger zone for an A380. (See below the section “Safety Analysis and Hazard Boundaries.”) Many factors could influence the vertical position of the wake danger zone. Specifically, if the vertical air mass motion and stratification are uncharacterized, they can move the wake up or down in a seemingly 

36 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY random fashion. There is also a large variation in the lifespan of a wake, depending on atmospheric turbulence (Crow, 1970). Due to the difficulty in obtaining all the parameters that affect these two dimensions of the wake location, the danger zone in these dimensions would be signifi- cantly larger than the nominal wake. On the other hand, there are fewer uncertainties in the lateral location of the wake. The wake danger zone can be shown on a two-dimensional mov- ing map display (Fiduccia, 2005). Most commercial aircraft now have a moving map display that could be used to display the wake danger zone associated with neighboring traffic within a certain altitude (e.g., 300 ft). This type of depiction might also be used on the radar displays used by ATC for all aircraft. ATC could then verify the safety of placing aircraft closer together in-trail when the predicted crosswinds indicate the safety of that separation. It would also enable ATC and pilots to verify the safety of more efficient spacing of aircraft on CSPAs (Powell et al., 2005). The wake danger zone can also be shown on a three-dimensional perspective display (synthetic vision), as shown by Holforty (2003). The ability to predict and display the wake of a neighboring aircraft has been demonstrated in flight experiments (Holforty and Powell, 2003), and the predicted wake location was found to be identical to the actual wake position. In this case, the parameters of the target (wake-generating) aircraft were well known and the winds were known and steady, mak- ing the accuracy of the wake location prediction most likely better than could be expected under more typical conditions. However, the wake visualization concept was proven and was judged by the participants as superior to today’s situation, where no information is available to pilots or controllers on the location of the wake. The flight experiments and computer studies cited above were not exhaustive, nor were they applied to commercial aircraft. There needs to be more research into the most useful display methods; verification and development of the models with larger aircraft; addition of airborne atmospheric, meteorological, and wake detection sensors as they become available; thorough error analyses to determine the model accuracy and corresponding danger zones; and enough demonstrations to establish acceptance by all stakeholders: pilots, controllers, operators, and the FAA. If accepted by all, the results of the research could enable reduced sepa- ration of aircraft in all phases of flight, with a level of safety equal to or better than today’s. This work would be most effectively conducted by a collaborative partnership between the FAA and NASA. Milestones for wake vortex visualization are listed in Table 3-2. 

TECHNICAL CHALLENGES 37 TABLE 3-2  Milestones for Wake Vortex Visualization Time Horizon Milestone Short term Review all real-time models and establish required measurements Establish error bounds for definition of wake danger zone Perform simulator studies of display options Medium term Use displays in simulator studies of CSPA procedures Flight test to validate depictions of the danger zone Long term Integrate airborne wake vortex sensors Certify sensors for use in ADS-B-equipped aircraft Use in flight test validation of CSPA procedures Finding 3-5. Onboard wake vortex visualization has been demonstrated in a proof-of-concept trial and can provide a safety net for dynamic spacing procedures. Recommendation 3-5. Wake vortex visualization concepts should be further explored and pursued. vortex alleviation Trailing vortices are an unavoidable consequence of finite-span lift- ing wings. The initial vortex strength scales with the aircraft weight and inversely with the aircraft wingspan. In general, vortex alleviation con- cepts aim at reducing the hazard posed by a vortex by altering the vor- tex’s characteristics. This is done by modifying the configuration of the source aircraft. Any such modifications must not significantly degrade the overall aircraft performance or they will not be viable for integration on the aircraft. There have been many attempts over the last 40 years to evaluate systems that provide some alleviation of the wake vortices in laboratory simulations. These systems are reviewed by Savaș (2005) and Crouch (2005). Vortex alleviation offers the potential to remove the vortex threat within a given distance of the generating aircraft. If successful, this solu- tion would allow following aircraft to fly closer to the “vortex-alleviated” aircraft. The benefits of alleviation would be available at any airport used by the vortex-alleviated aircraft. Vortex alleviation benefits are also expected to be weather independent; they may be more fully exploited because they can be counted on and forecast well in advance. However, the total benefit of vortex alleviation to the national airspace system will be dependent on the number and mix of aircraft that have incorporated alleviation techniques. Thus, even if a vortex-alleviation solution is found, 

38 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY it will take many years before the benefits can be fully realized, depending on the general applicability of the concept. The benefits of vortex alleviation can be measured by changes to the severity and the probability of vortex encounters. This requires mea- surements and/or predictions of the vortex characteristics as a function of distance behind the vortex-generating aircraft. Some form of vortex- encounter simulation can then be used to determine the potential severity of a vortex upset at a given distance behind the vortex-generating aircraft and to define the acceptable following distance enabled by each vortex alleviation concept. Implied in the assessment of vortex alleviation is a defined “hazard boundary” that can be used to judge if a given encounter is acceptable. The approaches to vortex alleviation can be grouped into two broad categories, one aimed at vortex modificationthat is, making vortices weaker or more diffuseand the other aimed at vortex breakupthat is, forming vortex rings. Vortex modification is achieved through alterations to the loading of the wingspan (and sometimes the horizontal tail). Vortex breakup is achieved through unsteady forcing, preferably using standard control surfaces. Vortex breakup relies on natural instabilities to amplify small vortex perturbations into large, wavy distortions of the vortices. The waviness ultimately leads to a periodic pinching of the vortices into vortex rings. The vortex rings undergo continued distortion, finally leading to a com- plete breakdown of the vortex system. The most promising way to accel- erate vortex breakup relies on the multiple-vortex-pair system that can be generated by aircraft with their flaps deployed (Crouch et al., 2001; Crouch, 2005). This approach uses conventional control surfaces to excite rapidly growing instabilities that can break the vortices into rings within distances of approximately 3 NM for a “heavy” sized aircraft—signifi- cantly shorter than the aircraft separation distances imposed today under IFR. However, because this system relies on specific features of the initial aircraft wake, it is not expected to be applicable to all aircraft. Also, the allowable following distance may be somewhat greater than the distance to pinching, depending on the specific criteria used for its determination and the specific aircraft pair. Some of the most promising approaches to vortex modification also make use of multiple-vortex-pair systems (Fabre et al., 2002; Durston et al., 2005; Savas, 2005). The addition of a strong vortex pair rotating counter to the basic wingtip vortices can lead to more diffuse vortices whose strength is the sum of the positive and negative vortices. However, the addition of the counter-rotating vortices costs aircraft performance. Even though it may dramatically alter the vortices, this approach does not look promising for implementation because of its effect on overall aircraft configuration. 

TECHNICAL CHALLENGES 39 Another approach that offers the potential for rapid distortion (breakup or pinching) of the vortices is to exploit transient growth or bypass mechanisms on the vortex or vortex pair (Pradeep and Hussain, 2006; Brion et al., 2007). Studies have shown significant alterations to the vortices, accelerating their decay, but the practical costs and value of this approach have not been assessed. There is very little vortex alleviation activity in the United States with no discernible effort at any of the federal agencies. Activity over the past decade has been concentrated in industry, in academia, and in Europe. Research in this area requires innovative concepts (which cannot be forecasted) and fundamental investigations into the effectiveness and viability of the concepts. It is essential that the viability of concepts be assessed so that the limited resources can be focused on the most promis- ing areas. Research on successful concepts should advance from compu- tational fluid dynamics (CFD) studies, to experimental demonstrations and on to an assessment within the complexities of the real atmospheric environment. Progress toward alleviation also requires advances in the definition of a hazard boundary that can be used to judge the alleviation concepts and estimate their benefit. Milestones for vortex alleviation are shown in Table 3-3. Finding 3-6. Vortex alleviation has the potential to significantly impact aircraft spacing requirements in the long term. Recommendation 3-6. Vortex alleviation ideas, including configuration changes and active and passive forcing, should be explored. Finding 3-7. NASA’s aeronautics program is well-aligned to conduct wake vortex alleviation work as medium- to long-term foundational research. TABLE 3-3  Milestones for Vortex Alleviation Time Horizon Milestone Short term High-level assessment of alleviation based on effectiveness and viability Multiple concepts explored Systems benefit studies conducted for best concepts Medium term Validation tests conducted for most viable concepts, at Reynolds numbers sufficient to represent an aircraft in flight, preferably via flight tests Assessment of performance under realistic atmospheric conditions for most viable concepts Long term Flight test conducted to demonstrate system effectiveness 

40 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY WEATHER FORECASTING Measurement of weather parameters relevant to wake vortex trans- lation and decay is discussed in the next section. However, to ultimately take advantage of dynamic spacing, it will be necessary to be able to forecast the effects on wake vortices far enough into the future to allow spacing of arriving aircraft according to the conditions that will exist at their approach time. It does no good if the arrival queue has already been established in the terminal area according to existing spacing criteria when vortex and weather measurements reveal that the spacing could have been safely reduced. To accomplish this, a local area forecast model of sufficient resolu- tion will be required and the data to initialize it will need to be available. Existing forecast models have been used with some success at STL to forecast wake transport. A higher-resolution model will be required to improve on this and to forecast wake vortex decay. The Weather Research and Forecasting (WRF) model currently under development by NCAR, NOAA, DOD, the FAA, the University of Oklahoma, and others has the potential to meet this requirement in the necessary time frame. It is expected to be field tested in 2009 (Skamarock et al., 2007). Data from a scanning lidar could be sufficient for initialization of the model but would be available only in clear weather. Other means for data acquisition in inclement weather exist, but they might not be adequate for this purpose, for example, providing eddy dissipation rates (EDRs). EDR is a quantita- tive measure of ambient turbulence used in calculations of vortex trans- port and decay. However, by the time EDR is needed, data ­transmitted by aircraft in flight may be sufficient. Software to compute EDR is now onboard United Airlines aircraft and is expected to be operational on two other U.S. carriers by 2009. When ADS-B is operational, these data should be available in near-real time. The WRF model will be capable of accepting data at sporadic intervals in time and space as they would be transmitted from aircraft. However, as of this writing, the model is still not capable of ingesting and processing EDR data, and no funding exists to develop this capability. Thus, although forecasts of conditions that would move wake vortices out of the areas of concern could be available in the needed time frame, forecasts of wake decay probably could not be achieved without further funding for model development. European research has demonstrated some short-term forecast model capability at Frankfurt (Gerz, 2007). NASA and the FAA have been part- ners in this effort and are therefore presumably aware of its characteristics and potential. European as well as U.S. WRF development efforts can be monitored for possible future U.S. application. Milestones for weather forecasting are shown in Table 3-4. 

TECHNICAL CHALLENGES 41 TABLE 3-4  Milestones for Weather Forecasting Time Horizon Milestone Short term Coordinate needs with NCAR Investigate possible data inputs for weather models Medium term Test integration of data and modeling Incorporate aircraft-gathered data into model Long term Certify weather prediction procedures Finding 3-8. The Weather Research and Forecasting model currently in development can provide the necessary weather prediction resolution to support the needs of NextGen. However, as presently conceived the model lacks the ability to incorporate eddy dissipation rates and thus will not provide the optimal set of parameters for predicting wake turbulence. Recommendation 3-7. Federal wake turbulence research should engage the Weather Research and Forecasting model effort in order to iden- tify the optimal set of weather parameters needed for predicting wake turbulence. Finding 3-9. European forecast models have demonstrated some short- term terminal area wake vortex forecast capability. Recommendation 3-8. Research should be done to ensure that weather modeling is adequate to predict wake vortex movement and decay. wake VORTEX modeling The initial (near-field) wake generated by an aircraft can be extremely complex, while the far-field wake (at, say, 3 to 8 miles) is a relatively simple vortex pair characterized by a circulation strength, a vortex span (distance between the vortices), a vortex core size, and some measure of the long-wavelength vortex distortion. The characteristics of the far-field wake are strongly influenced by atmospheric conditions. The position of the wake far behind the aircraft depends on wind conditions and the integrated history of the wake characteristics. Wake vortex modeling attempts to predict the basic characteristics of the vortices from the near field into the far field as a function of the generating aircraft and the ambi- ent atmospheric conditions. Wake vortex modeling plays a critical role in many concepts aimed at reducing aircraft IFR spacing requirements. First, vortex models could 

42 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY be used to compare the relative vortex strengths of different aircraft at a given following distance. This information could be used in conjunction with some form of hazard boundary to reevaluate the aircraft weight categories on which aircraft spacing is based. Second, if models could be used (in conjunction with local weather information) to predict the vortex positions with sufficiently small uncertainties, they could be used to visualize the vortices from the ground or from the cockpit. Finally, vortex models could enable the use of dynamic spacing based on local weather conditions. The simplest application is to use the model to iden- tify weather conditions in which vortex encounters can be completely avoided. Eventually, this could be extended to include weather condi- tions when vortices might be encountered but no longer pose a hazard. Different applications demand different levels of fidelity from the vortex models. Direct vortex measurement and detection can be used to supple- ment the vortex modeling for a dynamic spacing system, but the vortex models (in conjunction with local weather models) will determine how far in advance the airport capacity can be forecast. Vortex modeling can be broken into two somewhat distinct stages in the vortex development: wake initialization (near-field development) and wake evolution (mid- and far-field development). The wake-initialization phase provides a link between the generating aircraft and the basic wake characteristics needed to model the wake into the far field. This phase is relatively short, so atmospheric effects are neglected. This phase is often simplified by assuming an elliptic span loading (for the total aircraft lift); this provides the basic wake characteristics as a function of the aircraft weight, span, and speed. It is a very rough approximation for configura- tions with flaps deployed. For some applications (e.g., predicting vortex positions), the associated errors may not be significant to the far-field estimates. However, any relative comparison of specific-aircraft wake strengths will likely be very sensitive to the wake-initialization phase. This could prove to be important for the characterization of different aircraft wakes in the far field. The wake-evolution models start with a simple vortex pair and pro- vide a time history for wake development as a function of atmospheric conditions. The essential wake parameters for this part of the modeling are the vortex strength and the vortex span. The key atmospheric factors are the background turbulence, the stratification, and the lateral shear. Ground proximity is another key factor that significantly alters the devel- opment and is often treated as a distinct module in the modeling. Wake- evolution models are highly empirical; their validity is strongly tied to the quality of the data used in their development. Therefore, the parameters used in the models are dictated by measurement capabilities and are not necessarily the most physically relevant. The stochastic nature of the 

TECHNICAL CHALLENGES 43 atmospheric environment results in uncertainties in the model predic- tions; these uncertainties grow in time. They can overwhelm or amplify the uncertainties introduced with the wake-initialization model. Thus, any far-field predictions are probabilistic in nature, and this is specifically accounted for in some models. To advance modeling of the wake-initialization phase, quality flight data are needed at distances greater than approximately 15 wingspans, but not so far downstream that atmospheric influences overwhelm the ini- tial conditions. These data must be obtained under controlled conditions, where aircraft configuration details are varied and documented. The flight data could be supplemented with ground-based testing data if measure- ments are made sufficiently far downstream and the model represents a realistic aircraft configuration. These data can be used to assess current models or even to develop new models. Modeling of the wake-initialization phase has not received much attention at any of the federal agencies. Recent efforts are focused in industry and in European agencies and universities. The state-of-the-art approach is to establish a wake survey at the tail of the aircraft using wind-tunnel experiments or full-configuration CFD and then evolve this downstream with a code that is optimized for calculating the wake devel- opment (Czech et al., 2005; Winckelmans et al., 2005). The current models have been tested against a limited set of wind-tunnel data, but no direct comparisons to flight data have been made. To advance the prediction of wake evolution, the first challenge is to create probabilistic models that provide the nominal wake characteris- tics and the uncertainty associated with these values. These models can be used in conjunction with atmospheric measurements or predictions to predict the position of the wake as a function of time (or following distance). The combination of a probabilistic model and an atmospheric model can provide an “uncertainty box” around the predicted vortex positions that gives the range of possible positions within a given level of uncertainty. An indicator for how the uncertainty box grows with time as a function of different atmospheric conditions is needed to judge the utility of the predictions for a given application. The second challenge is understanding how the growth of uncertainties depends on the uncertain- ties in initial wake characteristics, and how this contributes to the scatter in typical flight data. Thus, current wake-evolution models would also benefit from quality flight data that could permit accurate initialization of the wake and permit comparison in the far field. The third challenge is estimating the overall confidence in the model under different conditions to assess when the models could be used and when real-time wake detec- tion would be required. Additional flight data will be required to advance the development of wake-evolution models. These data are needed to iso- 

44 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY late and model distinct atmospheric influences. Wake-evolution models may also benefit from more fundamental research on the effects of wind shear and atmospheric turbulence. Turbulence from the aircraft wake and the ambient environment can trigger vortex-core instabilities and bend- ing waves (Melander and Hussain, 1993), in addition to long-wavelength vortex pair instabilities (Crow, 1970). Modeling of the wake-evolution phase has been ongoing at NASA Langley Research Center, with research assistance from the Naval Post- graduate School and Northwest Research Associates. Real-time wake prediction models have been developed based on numerical models and validated with empirical data (Robins and Delisi, 2002). These real-time models use meteorological sensors to predict the movement and decay of the wake vortices and wake vortex measurement sensors to verify the results. Real-time models show the most promise for use in an operational environment because the codes are so efficient. Before real-time models can be used in air traffic management systems, they must be thoroughly validated. NASA’s AVOSS project, discussed above, used lidar to verify the predictions of its real-time model (O’Connor and Rutishauser, 2001). The NASA AVOSS Prediction Algorithm Model is a deterministic model (Proctor, 2007) based on the models of Sarpkaya (2000) and Greene (1986). A probabilistic version of the AVOSS model is under development (Proc- tor, 2007). Two well-known probabilistic models have been developed in Europe, the P2P model (Holzapfel, 2003) and the P-VFS model, which is based on the deterministic model VFS (Winckelmans et al., 2004). Mile- stones for wake vortex modeling are shown in Table 3-5. Finding 3-10. Wake vortex modeling is an essential element for most concepts aimed at reducing IFR spacing requirements. Finding 3-11. NASA’s aeronautics program is well-aligned to conduct medium- to long-term foundational wake vortex modeling. TABLE 3-5  Milestones for Wake Vortex Modeling Time Horizon Milestone Short term Flight test requirements defined based on modeling needs Probabilistic model developed based on the state of the art Vortex uncertainties predicted for different weather conditions Medium term Flight test conducted (as necessary) to support/validate modeling Wake models evaluated using flight data Range of applicability estimated for models Models applied to support recategorization efforts Long term Models applied to support dynamic spacing system 

TECHNICAL CHALLENGES 45 WAKE Vortex measurement Sensor technologies have contributed to the continued advancement of knowledge in wake vortex research since 1970 (Burnham, 1977). The role of these measurement systems has evolved as the technology capabil- ities advance to higher resolutions and longer ranges. For instance, some sensors are used to measure aircraft wake vortices, and others are used to measure the meteorological conditions for inputs into the wake vortex and weather prediction models. In addition, there are two fundamental roles that wake measurement systems serve: • Sensors for real-time models are required to detect the presence of aircraft-generated vortices in a volume of interest, typically in the approach or departure corridor. • High-resolution wake measurements are required to make high- resolution measurement of the velocity field sufficient to validate robust wake vortex models. The various sensing systems developed over the years have made a significant impact on wake vortex modeling and provided a better under- standing of wake vortex behavior. Current state-of-the-art sensor tech- nologies do not provide sufficiently detailed measurements of the three- dimensional aerodynamic phenomena needed by wake vortex ­modeling to advance the understanding of vortex behavior, and they cover only the landing and takeoff operations below 1 kilometer of aircraft altitude. Furthermore, the benchmark for wake sensors is an electro-optical sys- tem that works only in visual meteorological conditions. Although much progress has been made, there is still a need for higher resolution, higher spatial coverage, and all-weather capabilities. Other operational consid- erations include autonomous operation, real-time processing, all-weather performance, and cost (Burnham, 1997). The main challenges for the development of a sensor for real-time models are system reliability, all-weather operation, and spatial cov- erage. Because most wake vortex sensor systems were developed for research applications, continuous reliable operation was not an important design driver. From now on, however, system reliability will be a major design requirement since the eventual installation will be operating in the national airspace system. The second major challenge is the ability of the sensor system to operate in all weather conditions. Lidar systems are optical sensors that perform as long as the transmission of the laser is not interrupted. They have become the sensor of choice for measuring wake vortices, but they are unable to produce a better understanding of wake vortex behavior in all weather conditions, as required by an opera- tional system. The third challenge is to provide real-time detection of 

46 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY wake vortices and to monitor meteorological conditions along the flight path. Measured atmospheric parameters for locations along the flight path would improve the accuracy of a real-time wake vortex prediction system. If all of these challenges can be successfully overcome, they could contribute to the long-term vision whereby each aircraft acts as a sensor, providing data to a networked, high-resolution atmospheric monitoring network. While lidar systems are impractical for this purpose (they can detect wakes perpendicular to their line of sight but not parallel to it), if a suitable wake measurement concept became available, it could be added to an aircraft’s suite of sensors. Through some combination of modeling and monitoring, the aircraft would then be able to determine the location of its own wake, which could be added to the ADS-B information stream. Alternatively, an airborne sensor on each aircraft that measures the loca- tion of the wake of nearby traffic might be possible, eliminating the need for transmitting the information via ADS-B. This capability is related to a challenge discussed above in the section “Vortex Visualization: Cockpit and Controller.” Higher-resolution measurements would support the development of large eddy simulation (LES) of wake vortices and provide wake vortex modelers with the empirical data they need to validate their computa- tional models. Laser Doppler velocimetry—pulsed coherent and con- tinuous wave lidar—systems are able to detect and measure wake ­vortex characteristics from the ground. Other sensor systems such as radar acoustic sounding systems, monostatic acoustic vortex sensing systems, and windline anemometers have also demonstrated their capabilities. The measurement challenge is twofold. The first challenge is establishing the resolution requirements, which will require a collaborative effort between wake vortex modelers and sensor developers to avoid the ambiguities often associated with the research area. The second is developing a sensor that can make high-resolution measurements in the volume of interest. For takeoff and landing, most of the volume of interest is beyond the range of the higher-resolution continuous wave lidar. Other sensors are within range but unable to provide adequate data resolution. Wake vortex measurement systems have been developed with differ- ent principles of operation that take advantage of the particular meteoro- logical conditions in which they are designed to operate. Lidar sensors are electro-optical sensors that work best when the weather is clear. The prin- ciple of operation for lidar is measuring the backscatter of laser light from the particulates trapped in the vortices. Acoustic systems use ­ acoustic pulse to backscatter the vortex Doppler-shifted acoustic energy from the density variations in the vortices. Early wake vortex radar designs and simulations showed that water droplets trapped in the vortices would provide sufficient radar backscatter to detect and measure the vortices in 

TECHNICAL CHALLENGES 47 inclement weather. It would be very difficult to develop an all-weather sensor because the sensing principles vary for different weather condi- tions and are limited to the signal propagation capabilities of the particu- lar sensors. It is more reasonable to develop an all-weather system that includes multiple sensors. A lidar sensor can operate when the weather conditions are suited for its operation (e.g., visual meteorological condi- tions); when the weather conditions are better suited for an electro-mag- netic sensor that can penetrate and use the scattering properties of water droplets (e.g., instrument meteorological conditions such as rain, fog, and snow), a multisensor system can switch over to the most effective sensor. A multisensor systems approach would also be cost-effective, because to date, no single sensor has shown that it can provide detailed and reliable measurement of the wake vortices. Furthermore, the knowledge gained from the lidar sensor system can be used to advance the technology in the new inclement weather wake sensor. Wake vortex modelers can benefit from empirical data that can lead to the development of a validated wake decay model. Knowledge of how atmospheric turbulence influences the rate of wake vortex decay is as important as knowledge of how vortices are transported by the winds. The challenges include developing not only a validated wake decay model, but also technologies for measuring the volume of interest in takeoffs, cruise, and landings. Milestones for wake vortex measurement are shown in Table 3-6. Finding 3-12. Research and development of high-resolution wake vor- tex measurement sensors to support wake vortex modeling efforts has stalled since the late 1990s. TABLE 3-6  Milestones for Wake Vortex Measurement Time Horizon Milestone Short term Complete feasibility studies to develop high-resolution, all- weather wake vortex measurement system Develop functional requirements that are consistent with wake vortex modeling needs Medium term All-weather wake vortex measurement system Airborne wake vortex sensor Wake vortex measurement network Long term High-resolution, all-weather wake vortex measurement system High-resolution, all-weather wake vortex measurement network 

48 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY Finding 3-13. No high-resolution wake vortex measurement system capable of operating in inclement weather exists. Recommendation 3-9. Explore concepts for an all-weather, aircraft- based wake vortex measurement system that provides information on the location of the wake. Recommendation 3-10. Develop an all-weather wake vortex measure- ment system that provides high-resolution measurements of wake vor- tex characteristics sufficient to validate wake vortex modeling. safety analysis and hazard boundaries While in a perfect world means would be sought to avoid any wake vortex encounters, such avoidance is simply not possible, nor is it nec- essary for safety. A hazard boundary provides a demarcation between acceptable and unacceptable vortex encounters based on criteria devel- oped in conjunction with the pilot community. The maximum aircraft bank angle resulting from an encounter is the most common metric for characterizing the severity of an upset. An example of a simplified hazard boundary would be the level of maximum bank angle that would result in a given percentage of pilots executing a go-around. The definition of haz- ard boundaries, along with development of an end-to-end safety analysis methodology, is critical to any new system that is not based completely on avoidance. Finding 3-14. Although the current air transportation system was designed to avoid wake vortex encounters, they do occur and are safely tolerated using present spacing criteria. Finding 3-15. It is difficult to quantify acceptable reductions in wake turbulence spacing because there is no agreed metric for, nor definition of, hazard boundaries for wake encounters. Recommendation 3-11. A hazard boundary needs to be defined and used as a metric in forming spacing criteria. A rigorous safety analysis methodology is important for assessing the relative benefits of various mitigation solutions and for decision making in real time. The necessary elements of this methodology include (1) quan- tification of the uncertainties in wake prediction, aircraft characteristics, and local weather prediction; (2) assessment of how these uncertainties translate into a corresponding risk assessment; (3) the definition of an 

TECHNICAL CHALLENGES 49 acceptable level of risk and a corresponding definition of what constitutes a wake-turbulence hazard; and (4) a procedure for real-time decision making under uncertainty—for example, to determine dynamic spacing of aircraft. Strategies implemented to reduce the capacity constraints brought about by wake turbulence must not compromise safety. Concepts such as dynamic aircraft spacing, which have been proposed as long-term solutions to achieve NextGen capacity goals, clearly require explicit defi- nitions of hazard boundaries and acceptable levels of risk. This require- ment takes on more importance as the fleet mix comes to include a larger range of aircraft sizes—for example, VLJ and A380 aircraftthat move beyond the current size categories on which spacing is based. Hazard boundaries also play a role in evaluating proposed concepts for wake vortex alleviation. Previous work of this nature has been carried out as part of the Euro- pean S-Wake program under the Vortex Encounter Severity Assessment (VESA). VESA is a software tool developed by Airbus that estimates the probability of exceeding certain hazard levels given encounter conditions such as vortex characteristics, encounter geometry, and weather (Höhne, 2004). Ongoing work is addressing the validation of VESA based on flight test data. VESA is currently used in CREDOS, which is pursuing (with FAA and NASA participation) the development of crosswind reduced separations for departure. VESA is also used for safety assessment in the FLYSAFE and VITA projects, where parts of VESA are used to assess the severity of wakes. The challenge of hazard boundary definition is not being addressed by current research in the United States, but it must be addressed if the NextGen vision is to be achieved. The activities needed range from meth- odological and vehicle-level research to more operationally focused deter- mination of standards. This work would be most effectively conducted by a partnership between NASA and the FAA. Because defining a hazard standard is ultimately the responsibility of regulatory bodies, it should be led by the FAA. However, substantial research at the aircraft vehicle level is also required. For example, well-designed simulator studies are needed to characterize an acceptable wake encounter and to identify appropriate metrics for defining the hazard boundaries. This aspect of the work could be led by NASA, using contract support from aircraft manufacturers, air- lines, and pilots. The effort should build on work from VESA and include carefully constructed simulator studies using “naïve” airline pilots as well as representatives from the business and general aviation sectors. It is important that the study provide typical pilot reaction to unexpected encounters, and that the pilot subjects not be permitted more than a few encounters, so as to minimize learning. A wide range of typical aircraft 

50 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY TABLE 3-7  Milestones for Safety Analysis and Hazard Boundaries Time Horizon Milestone Short term Identify metrics for hazard definition Review European studies and complete detailed plan for simulator studies Begin conducting simulator studies Identify conservative hazard boundary Medium term Analyze results from simulator studies to quantify hazard Develop risk assessment methodology and apply it to simulator studies Refine hazard boundary based on available data Long term Test and implement refined hazard boundary Demonstrate real-time safety analysis in actual flight encounters must be explored, so that the envelope of the boundary can be reasonably defined with respect to leader and follower aircraft size and wake strength at encounter. The development of simulator software that could be shared with a number of airlines for use during their pilot train- ing programs would be one way to acquire useful data over a wide range of encounter situations. At the same time, research is needed to develop a methodology for probabilistic safety analysis. This is an important com- ponent that NASA could lead. Milestones for safety analysis and hazard boundaries are shown in Table 3-7. systems to gather data about wake events The challenge described here is to develop a means of collecting information from wake events that have actually occurred, as observed and reported by pilots. Currently only a very limited amount of informa- tion on a wake event is collected in any form by any agency in the United States, so that baseline data are not readily available. Without an event- driven database as a control, it will be very difficult to measure whether any future increase or decrease in wake events is the result of reduced spacing or is simply in line with current event levels. Pilots and controllers today do not have a simple system for report- ing the impact of a wake event in all phases of flight. The only wake reporting aid is the Supplemental Form, which must be attached to the NASA Aviation Safety Reporting System (ASRS) general form. The Type of Event/Situation section of the NASA general form (NASA, 1994) must be completed with the annotation, “Wake Turbulence.” The Supplemental Form is used only to report the reduced vertical separation minimums (RVSMs) in domestic airspace between 29,000 ft through 41,000 ft (flight 

TECHNICAL CHALLENGES 51 levels 290-410.) This limitation does not allow for reporting an event at lower altitudes, on arrival, approach, or departure. The FAA has funded a modification for the ASRS reporting system that would include wake encounters at lower altitudes but only on an official basis for those air- ports where current procedural mitigation work is under way (STL and IAH) and those airports where the A380 will operate (JFK and SFO). Many pilots are independently using the ASRS format to report wake encounters at altitudes other than those specified, but there needs to be a concerted effort to modify the system to officially collect the data from as many aircraft, in all flight phases, as possible. Establishing this baseline would accomplish several objectives. It would allow researchers to measure any future change in the number of events as a result of reduced aircraft spacing against the number of events given current aircraft spacing procedures. It would also begin an involvement with U.S. pilots and controllers at an early phase of research on the wake turbulence issue. This participation and communication will be crucial in gaining acceptance of any spacing modification as a result of the total wake turbulence research effort. Additionally, data collected in this effort could be utilized in establishing agreed-on metrics for defining hazard boundaries for the wake encounters. It will be important that data be collected for all phases of flight (departure, climb, cruise, descent, and approach) and that they reflect as many categories of aircraft as possible, from the A380 through light general-aviation aircraft. The ideal methodology for this effort should include data from both the leading and the trailing aircraft in an event. There are many ­parameters that would need to be identified in order to compile accurate and suffi- cient data, including aircraft types, altitudes, airspeeds, weights, and meteorological data, including temperature, wind direction, and velocity. Additionally, the location of the trailing aircraft relative to the leading aircraft must be known as accurately as possible. While the data might be readily available, their collection could prove to be rather problematic. Current-generation transport aircraft, including corporate and business aircraft, are equipped such that crew members could report most, if not all, of the data required. Flight data recorders could further provide infor- mation on any vertical acceleration and/or any change in attitude as a result of the event. Acquisition of the data from U.S.-certificated aircraft and crews will be the biggest challenge in this project. Any emphasis on reporting of events would undoubtedly increase the number of reports over the current level. Media interest in this higher number could cause both operators and manufacturers to become concerned about the liability, threatening the viability of a data acquisition program at its very outset. Therefore, a process for de-identification of data and restricted access to it would have to be in place before the start of the reporting. Even without 

52 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY access to information from flight data recorders, significant benefit could accrue from basic pilot reporting of a wake turbulence event. Any human- based wake reporting system will have to take into account human factors and the added workload. These factors will greatly affect the willingness to submit such data and their accuracy. Depending on the level of the data eventually obtained by this meth- odology, they could be used to correlate the operator input with detailed research data obtained from other elements of future wake turbulence research in the United States. Finding 3-16. Implementing a system to gather data on wake events in the short term could establish a baseline that could be used to quanti- tatively evaluate potential solutions, as well as gain the support from the operator and ATC community that will be necessary to increase system capacity. While the challenges of this data collection might sound ominous, there is precedent in the United States for such extracurricular reporting. In December 2004, the FAA issued an advisory circular (FAA, 2004) estab- lishing a methodology for reporting wildlife–aircraft strikes. The FAA, in cooperation with the Smithsonian Institution, gave pilots, maintenance personnel, and ground personnel detailed instructions for reporting a wildlife–aircraft strike; the methodology included transportation of bird feathers to the Smithsonian’s feather identification laboratory in Washing- ton, D.C. The FAA worked with the U.S. Air Force and the U.S. Depart- ment of Agriculture to improve the understanding and prevention of bird strike hazards. Through this program, the FAA maintains the National Wildlife Aircraft Strike Database and provides a very detailed form (FAA, 2006) for reporting such events that could serve as a model for reporting a wake turbulence event. Either the FAA or NASA could, if so directed, institute a wake tur- bulence event reporting system. The FAA effort could be an extension of the bird strike reporting activity, while the NASA effort would be an extension of the ASRS activity. Given the concerns about liability men- tioned above, it is useful that the ASRS already has de-identification and protection for reporting personnel. Flight personnel are already reporting to NASA without fear of enforcement action or liability. The question of access to flight data recorder information would remain problematic regardless of which agency was collecting the data. The plan for data collection would have three phases: • Phase OneA wake turbulence encounter database could be estab- lished. During this ramp-up phase, the aviation community would be 

TECHNICAL CHALLENGES 53 advised that a database was being established and reporting would be encouraged. At the same time both written and electronic means of report- ing the encounter would be developed. FAA air traffic controllers would be an important part of the data collection and therefore be included in the communication effort. For example, if the pilot of a trailing aircraft reported the encounter, the controller handling the flights would need to determine the data for the leading aircraft that could be generating the wake vortices. The research agency would establish a protocol for any follow-up (if required) to the initial report with both flight crew and controllers. In this phase of the program, the data collected could be used to develop flight simulator models of realistic wake vortex encounters. These models would be installed in the flight simulators of airlines and other commercial entities to better define hazard boundaries. Such scenar- ios would not only provide wake vortex encounter training for pilots, but could also be a real-life source of information to assist in defining hazard boundary metrics. This phase could be initiated in the short term. • Phase TwoData could continue to be collected and analyzed. Efforts could be started to reach agreement on access to flight data recorder information to supplement the basic report filed by pilots and controllers. Flight training and evaluation could continue. Input from line pilots to help in defining the hazard boundary could be gathered and analyzed. Because the issue of access to the flight data recorder arises at this point, Phase Two will probably not be attainable until the medium- to long-term time frame. • Phase ThreeAs any modifications to the separation requirements are implemented, the agency could then compare event frequency and impact data with similar data gathered before the separation require- ments were modified. If any degradation of safety were detected, separa- tion requirements could be modified accordingly. Flight simulator train- ing for wake vortex encounters could become part of a pilot’s initial and recurrent training and could be modified in concert with any advances or changes in wake turbulence alleviation. This phase could become an ongoing program much as the ASRS program is today. On the international scene, input from pilots and controllers on wake encounter events was being collected and processed on a very limited basis at London Heathrow airport (Critchley and Foot, 1991) as part of a larger, ongoing data collection effort in the United Kingdom. A sample wake turbulence encounter report form is reproduced in Appendix E. Milestones for systems to gather data about wake events are listed in Table 3-8. 

54 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY TABLE 3-8  Milestones for Systems to Gather Data About Wake Events Time Horizon Milestone Short term Undertake outreach efforts to aviation community Task data collection agency Establish reporting protocol Initiate reporting system and begin gathering data Interface with hazard definition effort Medium term Get agreement between operators and controllers on availability of flight recorder data Analyze reported data that will assist in safety analysis and determining hazard boundaries Long term Implement data-gathering system to evaluate changes to separation standards, such as from dynamic spacing systems and wake vortex alleviation Integrate wake vortex response into pilot training and modify it as required based on data analysis system-level study of benefits System-level studies are required to assess the relative benefits of wake turbulence mitigation strategies and to help with setting research priorities and using resources effectively. It is important that system-level studies cover a range of operational scenarios, weather scenarios, fleet mixes, and airport layouts. For example, identifying the potential ben- efits of reduced aircraft spacings in different operational settings requires first determining the relative impact of different options that may be applicable only at certain airports or under certain weather conditions. A number of potential improvements in wake vortex separation efficiency could be realized in the short and medium term that have been identified by the FAA. For reasons of airport geometry, terrain, fleet mix, or a com- bination of such factors, some of these improvements are applicable only to unique airport situations. Other wake vortex solutions are applicable to a particular set of airports or runway configurations, such as paral- lel staggered runways, parallel nonstaggered runways, and intersecting runways. These differences can become important, since (all other things being equal) solutions applicable to unique airport configurations have less overall potential for benefit than those applicable to a class of many airports, or those applicable generally. Another important aspect of system-level studies is to identify when and which other constraints in the system become active as the wake con- straints are modified. As discussed in Chapter 1, the aviation system is a complex system with multiple factors affecting system capacity. To realize 

TECHNICAL CHALLENGES 55 the full benefits of strategies for mitigating wake-turbulence constraints, these other factors must be studied and assessed. Research to address this challenge is being carried out by the JPDO Evaluation and Analysis Division (Borener, 2007). For example, a con- straints analysis has been carried out to examine and quantify the pri- mary factors limiting system performance. This analysis is being used to identify and quantify the long-pole factor(s), as well as to inform agency research and development plans to focus on key areas to help achieve NextGen goals. Ultimately the goal is to consider capacity, environment, safety, security, and costs, although initial analysis has been limited to capacity and environment. Another example is a study conducted to determine whether runway occupancy time or in-trail separation on final approach due to wake vortex constraints is the most binding constraint on capacity for SFO. The Ventana NextGen Portfolio Simulator is a quan- titative framework that takes performance estimates from these kinds of more detailed simulations and links them with a heuristic model to estimate the aggregate result of NextGen investments on system perfor- mance. Specifically, the tool is used to explore R&D portfolio design by quantifying how R&D leads to changes in the system performance over time as NextGen is implemented (Goldner and Borener, 2006). NASA Ames is pursuing development of the Airspace Concept Evalu- ation System (ACES) (Couluris et al., 2002; Sweet et al., 2002; NASA, 2005), a modeling and simulation environment for the national airspace system. ACES is intended to enable comprehensive assessment of the impact of new tools, concepts, and architectures. It covers aircraft flight from gate departure to arrival. The ACES environment is constructed so as to be flexible and extendable; that is, it supports plug-and-play assem- bly of different models. The program required to address this challenge would encompass both technical research to further develop the detailed simulation ­models described above and research that is more oriented to organization and man- agement to tie quantitative studies to research prioritization processes. Large-scale simulation and optimization methods play an important role in providing means to carry out system-level studies. The aviation system is extremely complex and has many uncertainties. Simulating the entire system at the level of detail required to assess the relative benefits of technological and operational solutions is a major challenge for the soft- ware development and modeling community. This challenge is described in Hunter et al. (2005) along with high-level requirements, architecture considerations, and software development strategies. The ongoing NASA and JPDO efforts described above have gone some way to addressing this challenge; continued focus, integration, and coordination in this area are essential. 

56 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY TABLE 3-9   Milestones for System-Level Study of Benefits Time Horizon Milestone Short term Simulation capabilities for arrival/departure operations extended to cover a range of operational scenarios, weather scenarios, fleet mixes, and airport layouts Medium term Fleetwide simulation capabilities for gate-to-gate operations Long term System-level simulation capability that includes the effects of uncertainty, to support dynamic spacing system design The system-level studies presented to the committee and documented in the literature have been for the most part limited to studies at specific airports (e.g., the runway occupancy versus in-trail separation study was done at SFO). The challenge is so important that resources need to be directed toward extending simulation capabilities to cover a range of operational and weather scenarios, fleet mixes, and airport layouts. Mile- stones for system-level study of benefits are shown in Table 3-9. Finding 3-17. System-level studies are an essential element of a wake turbulence research program. They are basic to ensuring that (1) research priorities will be set in a rational manner, (2) the actual realizable ben- efits of wake turbulence solutions will be known, (3) key constraints can be identified, and (4) NextGen capacity goals can be achieved. Recommendation 3-12. The current JPDO research in system-level modeling of the air transportation system should be continued and resources should be directed to extending simulation capabilities to cover a range of operational scenarios, weather scenarios, fleet mixes, and airport layouts. References Borener, S. 2007. Reduced wake vortex and runway occupancy time analysis, presentation to the Committee to Conduct an Independent Assessment of the Nation’s Wake Turbu- lence Research and Development Program, April 2, 2007, Washington, D.C. Brion, V., D. Sipp, and L. Jacquin. 2007. Optimal amplification of the Crow instability. Physics of Fluids 19:111703, doi:10.1063/1.2793146. Bryant, W., and P. Fiduccia. 2005. FAA/NASA Wake Turbulence Research Program Wake- VAS ConOps Evaluation Report: Phase III Elements: Far Term Candidate Operational Enhancements, Ground-Based and Airborne Enhancements to Phase II ConOps. NASA Langley Research Center. Hampton, Va. July 11. Burnham, D.C. 1977. Review of vortex sensor development since 1970. Proceedings of the Aircraft Wake Vortices Conference, DOT Transportation Systems Center. FAA-RD-77- 68. June, pp. 47-66. 

TECHNICAL CHALLENGES 57 Burnham, D.C. 1997. Ground-based wake vortex sensor technology: Current capabilities, future prospects. TP 13166. Proceedings of the International Wake Vortex Meeting. H. Posluns, ed. Transport Canada, Montreal, Ottawa, December, pp. 107-108. Couluris, G.J., C.G. Hunter, M. Blake, K. Roth, D. Sweet, P. Stassart, J. Phillips, and A. Huang. 2002. National Airspace System simulation capturing the interactions of air traffic management and flight trajectories. AIAA 2003-5597. AIAA Modeling and Simulation Technology Conference, August, Monterey, Calif. Critchley, J.B., and P.B. Foot. 1991.United Kingdom Civil Aviation Authority wake vortex database: Analysis of incidents reported between 1982 and 1990. CAA Paper 91015. Crouch, J.D. 2005. Airplane trailing vortices and their control. Comptes Rendus Physique ������������������������ 6:487-499. Crouch, J.D., G.D. Miller, and P.R. Spalart. 2001. Active-control system for breakup of air- ����������������������������������������������� plane trailing vortices. AIAA Journal 39:2374-2381. Crow, S. 1970. Stability theory for a pair of trailing vortices. AIAA Journal 8:2172-2179. Czech, M., G. Miller, J. Crouch, and M. Strelets. 2005. Predicting the near-field evolution of airplane trailing vortices. Comptes Rendus Physique 6:451-466. Durston, D.A., S.M. Walker, D.M. Driver, S.C. Smith, and Ö. Savaș. 2005. Wake-vortex al- leviation flowfield studies. Journal of Aircraft 42:894-907. Erikson, P. 2007. Wake vortex research in Europe, presentation to the Committee to Conduct an Independent Assessment of the Nation’s Wake Turbulence Research and Develop- ment Program, April 2, 2007, Washington, D.C. FAA (Federal Aviation Administration). 2004. Reporting Wildlife Aircraft Strikes. Advi- sory Circular 150/5200-32A. Available at <www.faa.gov/airports_airtraffic/airports/­ resources/advisory_circulars/media/150-5200-32a/150_5200_32a.pdf>. FAA. 2006. Bird/Other Wildlife Strike Report. FAA Form 5200-7 (11-97). Available at <forms. faa.gov/forms/faa5200-7.pdf>. Fabre, D., L. Jacquin, and A. Loof. 2002. Optimal perturbations in a four-vortex aircraft wake in counter-rotating configuration. Journal of Fluid Mechanics 451:319-328. Fiduccia, P. 2005. Recommendations and Conclusions from the Concept Evaluation Team Report, presentation at the WakeNet-USA Meeting, Boca-Raton, Florida, March 16- 17, 2005. Available at <http://wwwe.onecert.fr/projets/WakeNet2-Europe/fichiers/ pastEvents2005/bocaRaton/B06-CONOPS%20Evaluation%20%5BFiducia%5D.pdf>. GAO (Government Accountability Office). 2003. Aviation Infrastructure: Challenges Related to Building Runways and Actions to Address Them. Report to the House Subcommit- tee on Aviation, House Committee on Transportation and Infrastructure. GAO-03-164. Available at <www.gao.gov/new.items/d03164.pdf>. Gerz, T. 2007. Wake vortex research at Deutsches Zentrum für Luft- und Raumfahrt, presen- tation to the Committee to Conduct an Independent Assessment of the Nation’s Wake Turbulence Research and Development Program, May 15, 2007, Washington, D.C. Goldner, D.R., and S.S. Borener. 2006. Evaluating NGATS Research Priorities at JPDO. AIAA 2006-7726. 6th AIAA Aviation Technology, Integration and Operations Conference (ATIO), September 25-27, Wichita, Kans. Greene, G.C. 1986. An approximate model of vortex decay in the atmosphere. Journal of Aircraft 23:566-573. Hahn, K., C. Schwarz, and H. Friehmelt. 2004. A simplified hazard area prediction (shape) model for wake vortex encounter avoidance. ICAS 2004. 24th Congress of the Aero- nautical Sciences. Höhne, G. 2004. Vortex encounter severity assessmentVESA. Presented at WakeNet 2 Europe Workshop, Hamburg, Germany, May 11. Holforty, W.L. 2003. Flight Deck Display of Neighboring Wake Vortices. Ph.D. dissertation. Stanford University. 

58 WAKE TURBULENCE—AN OBSTACLE TO INCREASED AIR TRAFFIC CAPACITY Holforty, W.L., and J.D. Powell. 2003. Airborne wake avoidance and visualization experi- ment: Evaluating a wake vortex display. Society of Flight Test Engineers Symposium, September, Portsmouth, Va. Holzapfel, F. 2003. Probabilistic two-phase wake vortex decay and transport model. Journal of Aircraft 40:323-331. Hunter, G., K. Ramamoorthy, P. Cobb, A. Huang, M. Blake, and A. Klein. 2005. Evaluation of future national airspace system architectures. AIAA 2005-6492. Modeling and Simula- tion Technologies Conference and Exhibit, August 15-18, San Francisco, Calif. JPDO (Joint Planning and Development Office). 2007. Concept of operations for the Next Generation Air Transportation System, version 2.0. June 13. Available at jpdo.gov/­ library/NextGen_v2.0.pdf. Last accessed August 2, 2007. Lang, S. 2007. Wake turbulence research management plan, presentation to the Committee to Conduct an Independent Assessment of the Nation’s Wake Turbulence Research and Development Program, April 2, 2007, Washington, D.C. Lunsford, C., A. Mundra, L. Audenaerd, J. Cheng, C. Devlin, A. Gross, R. Mayer, J. Sherry, W. Bryant, E. Johnson, and B. McKissick. 2005. Wind dependent concepts for wake avoidance: A comparative analysis of capacity benefits and implementation risks. 24th Digital Avionics Systems Conference, October 30. Ma, K.L., and K.C. Zheng. 1994. 3D visualization of unsteady 2D airplane wake vortices. Proceedings of IEEE Visualization ’94, Washington, D.C. October 17-24, pp. 124-131. Melander, M.V., and F. Hussain. 1993. Coupling between a coherent structure and fine-scale turbulence. Physical Review E. 48:2669-2689. NASA (National Aeronautics and Space Administration). 1994. General form. NASA ARC 277B. Available at <asrs.arc.nasa.gov/forms/PDF_Files/general.pdf>. NASA. 2005. Airspace Concept Evaluation System. Available at <vams.arc.nasa.gov/­ activities/aces.html>. O’Connor, C.J., and D.K. Rutishauser. 2001. Enhanced Airport Capacity Through Safe, Dynamic Reductions in Aircraft Separation: NASA’s Aircraft Vortex Spacing System (AVOSS). NASA TM-2001-211052. Available at <ntrs.nasa.gov/archive/nasa/casi.ntrs. nasa.gov/20010089337_2001145219.pdf>. Last accessed August 2, 2007. Powell, J.D., C.W. Jennings, and W.L. Holforty. 2005. Use of ADS-B and perspective displays to enhance airport capacity. Proceedings of 24th IEEE Digital Avionics Systems Confer- ence, October, Washington, D.C. Pradeep, D.S., and F. Hussain. 2006. Transient growth of perturbations in a vortex column. Journal of Fluid Mechanics 550:251-288. Proctor, F. 2007. Modeling of atmospheric and ground effects on aircraft wake vortices, presentation to the Committee to Conduct an Independent Assessment of the Nation’s ��������������������������������������������������������������� Wake Turbulence Research and Development Program, May 15, 2007, Washington, �������������������������� D.C. REDAC SSWG (Research, Engineering and Development Advisory Committee Separation Standards Working Group). 2006. Final Report. September 20. Robins, R.E., and D.P. Delisi. 2002. NWRA AVOSS Wake Vortex Prediction Algorithm Version 3.1.1, NASA/CR-2002-211746, June. Northwest Research Associates, Bellevue, Wash. Sarpkaya, T. 2000. A new model for vortex decay in the atmosphere. Journal of Aircraft 37:53-61. Savaș, Ö. 2005. Experimental investigations on wake vortices and their alleviation. Comptes Rendus Physique 6:415-429. Skamarock, W.A., J.B. Klemp, J. Dudia, D.O. Gill, D.M. Barker, W. Wang, and J.G. Powers. 2007. A description of the advanced research WRF, Version 2. NCAR Technical Note NCAR/TN-468+STR. Available at <www.mmm.ucar.edu/wrf/users/docs/arw_v2_ bw.pdf>. 

TECHNICAL CHALLENGES 59 Sweet, D.N., V. Manikonda, J.S. Aronson, K. Roth, and M. Blake. 2002. Fast-time simulation system for analysis of advanced air transportation concepts. AIAA-2002-4593. AIAA Modeling and Simulation Technology Conference, August, Monterey, Calif. Winckelmans, G., V. Treve, and O. Desenfans. 2004. Real-time safety advising system in re- duced wake vortex separations operation, in 2nd WakeNet2-Europe Workshop: ­Capacity Gains as a Function of Weather and Weather Prediction Capabilities, ­Langen. Winckelmans, G., R. Coclé, L. Dufresne, and R. Capart. 2005. Vortex methods and their a ­ pplication to trailing wake vortex simulations. Comptes Rendus Physique 6:467-486.