Environmental Optimization of Aircraft Departures: Fuel Burn, Emissions, and Noise (2013)

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.

72 C-1. Introduction With the increase in traffic in the National Airspace Sys- tem (NAS), the effects of any changes to air traffic procedures on the capacity of airports must be considered, as airports are one of the major areas of congestion in the NAS. If new proposed procedures decrease the capacity of an airport, a trade-off must be done to examine the other benefits from the procedure and to determine if those benefits outweigh the cost of decreasing capacity. Decreased environmental impacts are some of the benefit that has resulted in decrease in capac- ity due to restrictions on flight paths of aircraft into and out of airports. Noise Abatement Departure Procedures (NADPs) historically have resulted in decreases of capacity because they normally restrict where aircraft can fly around airports. However, with the development of quieter engines and air- craft, new NADPs may be able to gain these losses in capac- ity back. Because aircraft now create less noise, new NADPs could allow for a wider range of flight paths around airports while still maintaining low noise levels. By designing NADPs to balance the need for increased efficiency but less noise, fuel burn, and emissions, optimal procedures can be created that can actually improve an airport’s efficiency. When perform- ing trade-offs on airport efficiency, calculating the airport capacity allows any changes in efficiency to be quantified. The airport capacity is shown as a Pareto frontier, as seen in Figure C-1, which describes the maximum departure rate for a given arrival rate under a given set of conditions. These conditions include the weather, which affects the airport con- figuration and the flight rules being used, and the type of air- craft being considered. The standard method for calculating the capacity of an airport is to pick a set of conditions and then use the FAA procedures for the terminal area to deter- mine how many departures can leave for a given arrival rate. The arrival rate is varied so as to calculate the entire boundary, while still ensuring the required minimums are maintained. This document will describe the methods used for calculating the capacity of a single runway at an airport and then apply those principles to calculating the capacity of an entire airport. As seen in Figure C-1, the capacity curve is only the theo- retical description of the relation between arrival and depar- ture rates. Very often these rates are not achieved due to less than optimal conditions or not enough traffic to maximize the airport’s resource. In Figure C-1, most reported rates occurred at well below the calculated Pareto frontier. How- ever, the actual rates can exceed the calculated capacity curve, as seen in Figure C-1 as the few data points outside of the curve. This can indicate over-utilization of the airport or a high level of optimization by controllers and airlines. C-2. Single Runway Capacity C-2.1. Overview When calculating the capacity of a single runway, a set of initial conditions must first be chosen. This set of conditions includes the fleet mix using this runway and the type of flight rules being used. Once the fleet mix has been determined, arrival and departure sequences can be created. The arrival sequence is spaced out and threshold crossing times are cre- ated. Once these times and spacings have been determined, the arrival and departure sequences are iterated through allowing a departure to take off every time that there is a large enough gap between arrivals. Once the arrival sequence has been completely iterated through, the arrival and depar- ture rates are calculated based on the total number of arrivals and departures that occurred in the time it took for all of the arrivals to land. The spacing is then varied to determine the departure rates at varying arrival rates. C-2.2. Initial Conditions The two major initial conditions that influence the capacity of a runway are the flight rules being applied and the fleet mix A P P E N D I X C Capacity Modeling Protocol

73 used on the runway under consideration. Whether aircraft are being flown under Visual Flight Rules (VFR) or Instru- ment Flight Rules (IFR) greatly affects the capacity of a run- way and by extension an airport. The decision between IFR and VFR is determined by the weather conditions. Normally, when calculating capacity, instead of setting the weather con- ditions, the set of flight rules being used is decided upon. For example, when the FAA performed their benchmark of the capacity of airports in 2004, they picked three sets of con- ditions to examine each airport under. First, they considered the case of optimal conditions, where visual approaches were used. Then, they calculated the capacity under marginal con- ditions with instrument approaches but visual separation was still maintained. Finally, they determined the capacity under IFR conditions. The other set of initial conditions that must be initialized is the fleet mix, which determines the sequence of aircraft types for the runway. The types of aircraft can be simplified into the four types of aircraft listed in the FAA regulations. These types are small, large, Boeing 757, and heavy. Further detail is not needed as all separation requirements can be determined from these types. These sequences can be set in a variety of ways. A sequence can be created by determining the percentages of each type of aircraft that frequent the airport under consideration. These percentages can then be used along with a uniform random number generator to create a random sequence with the correct percentages of aircraft types. Another way to initialize each sequence is to use true sequences observed at the airport. C-2.3. Set Arrival Times The next step in calculating the capacity is to set the times at which each arrival in a sequence will arrive at the specified runway. This process is done by applying the FAA required spacing. The minimum IFR separation between aircraft in an arrival sequence due to wake vortices is specified in FAA Order JO 7110.65S, Para 5-5-4. These minimum separations are described below in Table C-1. The minimum separation between aircraft when apply- ing visual separation is not specified by the FAA. Pilots are given latitude to maintain a safe separation between each other. As a result, the visual separations tend to be smaller. The observed average separation between aircraft is shown in Table C-2. These separation distances are the distance between the two arrivals when the leading arrival is crossing the runway threshold. Another restriction on whether or not an arrival can land is that the runway must be clear. To take this into account, the observed runway occupancy times in Table C-3 can be used. Figure C-1. FAA reported capacity curve for Reagan Washington National Airport with actual data shown from Airport Capacity Benchmark Report 2004. Trailing Aircraft Leading Aircraft (nm) Small Large B757 Heavy Small 2.5 2.5 2.5 2.5 Large 4 2.5 2.5 2.5 B757 4 4 4 4 Heavy 6 5 5 4 Table C-1. IFR minimum wake vortex separation between arrivals.

74 The arrival time for each aircraft in the arrival queue is said to be the time at which the aircraft crosses the runway threshold. To determine the arrival time of each arrival, the previous requirements are applied to the arrival sequence. Because the separations are in terms of distances, they must be converted to times. In order to do so, the average approach speeds in Table C-4 are used to convert the dis- tances into times. In addition to the calculated arrival times, an additional time is added to each arrival time to allow for varying the arrival rate. This additional time is found from a random number generator with an exponential distribution defined by f(x) = le-lx where f(x) is the probability density function and x is the random variable, which is in this case time. This distribution is varied to create a range of arrival rates. Thus, when l approaches infinity, the resulting times from the dis- tribution approach zero and so the maximum arrival rate is obtained. The departure times are not set initially. It is not known when a departure will be able to take off until the related arrival and departure sequences are set. Once the airport configuration and flight rules have been set and the arrival times calculated, all of the arrival and departure sequences must be iterated through following the FAA procedures according to first in/first out. The arrivals cross the threshold at the set arrival times and as many as possible departures are allowed to take off. The processing stops as soon as the arrival sequence is iterated to the end. Once this happens, the number of arrivals is summed to determine the total number of arrivals and then divided by final arrival time to calculate the arrival rate. The same calculations are done on the departures to determine the departure rate. The departure sequence is assumed to be always full so that the limit on departure rate is not dependent on a lack of departures in a sequence. Therefore, if a departure sequence is ever emptied during the processing, additional departures must be added using whatever method was used to generate the original sequence or a departure can be generated during the processing each time a departure is needed. The arrival sequences must be sufficiently long so as to remove noise due to the randomization, or the process- ing must be repeated multiple times for a given exponen- tial distribution in order to average out the noise. Once a departure rate has been determined for a given exponential distribution and thus a given arrival rate, the process must be repeated with a different distribution. In this way, the entire capacity curve can be calculated for an entire range of arrival rates. The rules governing when a departure can take off are described in FAA Order JO 7110.65S, Section 9. When exam- ining Noise Abatement Departure Procedures (NADPs), the key regulations involved deal with the spacing between departures. A departure is allowed to begin takeoff roll once the preceding departure “has crossed the runway end or has turned to avert any conflict [JO 7110.65S 3-9-6.a]” as shown in the diagram in Figure C-2. However, if distances can be determined, the preceding departure needs to only be airborne and a minimum dis- tance from the current departure. The minimum distances are listed in Table C-5. The categories used in the table are defined as follows: Category I are small single propeller driven aircraft weighing less than 12,500 lbs.; Category II are small twin engine propeller driven aircraft weighing less than 12,500 lbs.; Category III are all other aircraft. The arrival times must also be taken into account when determining if a departure can take off. Because the arrival times are fixed, if a departure is allowed to take off, any Trailing Aircraft Leading Aircraft (nm) Small Large B757 Heavy Small 1.9 1.9 - 1.9 Large 2.7 1.9 - 1.9 B757 - - - - Heavy 4.5 3.6 - 2.7 Table C-2. Observed average separation between visually separated arrivals. Aircraft Type Occupancy Time (sec) Small Large B757 Heavy 50 60 60 70 Table C-3. Average arrival runway occupancy time. Aircraft Type Average Speed (knots) Small Large B757 Heavy 90 130 130 150 Table C-4. Average arrival speed. Figure C-2. Same runway separation.

75 spacing requirements between arrivals and departures must be satisfied. For an arrival to land, the previous depar- ture on the runway must have crossed the runway end (JO 7110.65S 3-10-3.a.2). However, if distances can be deter- mined, the distances in Table C-5 can be applied where the leading aircraft is the departure and the trailing aircraft is the arrival. To determine how long it will take a departure to cross the end of the runway, the following assumptions can be made about the departure profile. The departure can be assumed to accelerate linearly for 5000 ft. down the runway until it reaches its average departure speed shown in Table C-6. After that the departure can be assumed to maintain a ground speed equal to its average departure speed. Other restrictions are based on wake turbulence constraints. An IFR/VFR aircraft departing behind a heavy jet/B757 must be separated by 2 minutes, when both are departing from the same runway. If a departure follows a heavy/B757 arrival or an arrival follows a heavy/B757 departure, these aircraft must be separated by 2 minutes if they are on a runway with a displaced landing threshold if the projected flight paths will cross. Small aircraft must be separated from “a large aircraft taking off or making a low/missed approach when utilizing opposite direc- tion takeoffs on the same runway by 3 minutes unless a pilot has initiated a request to deviate from the 3 minute interval (JO 7110.65S 3-9-6.i).” All aircraft departing “behind a heavy jet /B757 departing or making a low/missed approach when utilizing opposite direction takeoffs or landings on the same or parallel runways separated by less than 2,500 ft. (must be separated by) 3 minutes (JO 7110.65S 3-9-6.j).” C-3. Multi-Runway System Capacity C-3.1. Overview The capacity of a multi-runway system (i.e. an airport) is determined in a similar manner to that of a single runway. However, because there are multiple runways, events on one runway can affect events on other runways. Yet, the steps for calculating the capacity remain the same. The initial condi- tions must be set, arrival and departure sequences created and then the sequences iterated through using FAA proce- dures to determine when events can occur. C-3.2. Initial Conditions In addition to the type of flight rules being applied, when calculating the capacity of an airport, the runway configu- ration must also be decided. The majority of airports have multiple configurations that are used depending on the winds and visibility. The capacity of an airport is calculated with respect to a specified runway configuration and will change depending on which runways are in use. Also, the number of aircraft sequences that need to be generated will also vary depending on the runway configura- tion being considered. For every runway being used, at least one sequence must be generated. If a runway is being used for both arrivals and departures, a sequence for both must be created. However, if a runway is dedicated to arrivals or departures, only one sequence needs to be generated. For example, a single runway handling both arrivals and depar- tures requires the same number of sequences as two parallel runways with one runway dedicated to arrivals and the other to departures. C-3.3. Set Arrival Times With multiple runways, separate arrival sequences interact with each other as arrivals on intersecting runways restrict each other. These interactions are described in JO 7110.65S 3-10-4. These procedures can be summarized to say that an arrival cannot land until an arrival or departure on an intersecting runway or flight path has already crossed the intersection or has turned to avoid crossing the path of the arrival, or is stopping and holding short of the intersection. Wake vortex restrictions also apply in this case as well. If an arrival is going to fly through an intersection where a heavy/ B757 has just flown through as it departs, the arrival must be separated from the heavy/B757 crossing the intersection by 2 minutes. In order to determine when an arrival will cross an inter- section, a general landing pattern can be assumed. The arrival pattern is simplified to having the arrival cross the threshold at an elevation of 50 ft., touch down on the runway 1000 ft. beyond the threshold and then decelerate from its average arrival speed listed in Table C-4 to a stop 5000 ft. down the runway. Between crossing the threshold, the arrival can be assumed to perform a quarter g pull up. Table C-5. Minimum distance between departing aircraft. Trailing Aircraft Leading Aircraft (ft.) Category I Category II Category III Category I 3000 4500 6000 Category II 3000 4500 6000 Category III 6000 6000 6000 Table C-6. Average departure speed. Aircraft Type Average Speed (knots) Small Large B757 Heavy 100 150 150 170

76 C-3.4. Departure Spacing If aircraft are on not on the same runway, but are on inter- secting runways, the departure may not begin takeoff roll until the following conditions are met. If the preceding air- craft is another departure, the current departure cannot begin its take off roll until the preceding departure has “passed the intersection, has crossed the departure runways, or is turning to avert any conflict (JO 7110.65S 3-9-8.b.1).” If the preced- ing aircraft is an arrival, the departure cannot begin rolling until the “preceding arriving aircraft is clear of the landing runway, completed the landing roll and will hold short of the intersection, passed the intersection, or has crossed over the departure runway (JO 7110.65S 3-9-8.b.2).” Other restrictions are based on wake turbulence constraints. An IFR/VFR aircraft departing behind a heavy jet/B757 must be separated by 2 minutes, when both are departing from the same runway or parallel runways separated by less than 2,500 ft. If a departure follows a heavy/B757 arrival or an arrival follows a heavy/B757 departure, these aircraft must be separated by 2 minutes if they are on a runway with a displaced landing threshold if the projected flight paths will cross. Small aircraft must be separated from “a large aircraft taking off or making a low/missed approach when utilizing opposite direc- tion takeoffs on the same runway by 3 minutes unless a pilot has initiated a request to deviate from the 3 minute interval (JO 7110.65S 3-9-6.i).” All aircraft departing “behind a heavy jet /B757 departing or making a low/missed approach when utilizing opposite direction takeoffs or landings on the same or parallel runways separated by less than 2,500 ft. (must be separated by) 3 minutes (JO 7110.65S 3-9-6.j).” Other wake turbulence constraints must be used when aircraft are on intersecting runways or flight paths. An IFR/ VFR departure must be separated from a departing heavy/ B757 by 2 minutes when the preceding departure is on an intersecting runway and the projected flight paths will cross or the two aircraft are on parallel runways separated by more than 2,500 ft. if the projected flight paths will cross. Also, in the case of an arriving heavy/B757 landing on a crossing run- way in front of a departing IFR/FR aircraft, the departing air- craft must be separated by two minutes if the departure will fly through the airborne path of the preceding arrival (JO 7110.65S 3-9-8.b.3-4). C-4. Results The results of these calculations should be a curve simi- lar to that shown in Figure C-1 or below in Figure C-3. The departure rate is expected to drop off with an increase in arrival rate and vice versa. The exception rule is the case of Figure C-3. Example of calculated runway capacity Pareto frontier.

77 independent runways independent of each other being used for arrivals and departures. In this case, the expected result is a constant departure rate for any arrival rate since the depar- ture rate does not depend on the arrival rate. The calculated arrival and departure rates need not go to zero as seen in Figure C-1. The end point at the maximum arrival rate is due to the fact that there is a maximum arrival rate that certain configurations can handle. So for this case, even if there were no departures, the arrival rate could not increase any more. The other end point in this case is a result of the maximum lambda describing the distribution for the spacing between arrivals. This shaping variable could be increased further so as to decrease the arrival rate and con- tinue the curve downward.