AIAA AVIATION Forum 10.2514/6.

2022-4015
June 27-July 1, 2022, Chicago, IL & Virtual
AIAA AVIATION 2022 Forum

Flight Loads Analysis of CL-415 Scoopers

Linda K. Kliment 1 and Kamran Rokhsaz 2
Wichita State University, Wichita, KS 67260-0042, USA

John A. Nelson 3
United States Forest Service, Boise, ID 83705, USA
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

Flight data recorded between 2015 and 2019 on a fleet of four CL-415 aircraft has been
analyzed for the purpose of developing flight loads spectra. A low-pass filter has been used to
remove the effects of local and structural vibrations from the recorded vertical load factors.
Vertical load factors have been separated into gust and maneuver categories. Frequency of
occurrence of the vertical load factors has been determined and expressed in the form of
exceedance spectra for various altitude bands. The frequency of occurrence of the load factors
has been shown to correlate well with altitude above ground level. The results have been
compared with those of four other operations; single-engine-air tankers, ASM/leads, BAe-
146/RJ-85 large air tankers, and a limited amount of agricultural spraying. The magnitude
and the frequency of the load factors have been shown to be larger than all but those of
ASM/lead flights.

I. Introduction

S INCE mid-2000’s, the United States Forest Service (USFS) has been requiring all aircraft operating under contract
to be equipped with Digital Flight Data Recorders (DFDRs). The recorded data, that is proprietary to the operators,
is stored in a central repository. By contract, the data is made available to the USFS in its quest for continuous health
monitoring of the fleet.
Wichita State University (WSU) is given access to this data for determining the aircraft operational flight loads
spectra. The results can be used by the Original Equipment Manufacturers (OEMs) and the operators to refine the
aircraft limitations or maintenance and inspection schedules. If needed, these results can also be used by the regulating
agencies to refine the standards governing such operations. Some examples of previous analyses can be found in Ref.
1-4.
The legacy large air tanker fleet consisted mostly of reconfigured military aircraft. Following a 2012 report [5] by
the USFS and the Department of Interior, the operators gradually replaced the legacy air tanker fleet with newer,
faster, and more cost-effective aircraft. The current fleet consists mostly of civil transport aircraft modified as air
tankers. One of the newer additions to the fleet, although not in the “large” air tanker category, is the Bombardier CL-
415 Super Scooper. Unlike other air tankers, this amphibious aircraft was designed specifically for aerial firefighting.
The focus of the present article is the operational loads experienced by this fleet. While these aircraft differ from
heavy air tankers substantially in size and payload, their can deliver as much or more water per mission to a fire site.
For the purpose of the present analysis, flight loads data was gathered on four airframes operating for several years,
resulting in a large volume of data. The recorded data spanned from early 2015 through the end of 2019 calendar
years.

1
Associate Professor, Department of Aerospace Engineering, Senior Member AIAA.
2
Professor Emeritus, Department of Aerospace Engineering, Associate Fellow AIAA.
3
Airworthiness Branch Chief, Fire and Aviation Management.

Copyright © 2022 by Linda K. Kliment, Kamran Rokhsaz, and John A. Nelson. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.
 II. Method of Analysis

A. Aircraft Description
Some aircraft characteristics, dimensions, and limitations are shown in Table 1. Most of these values are from
Type Certificate Datasheet [6] and Aircraft Flight Manual [7]. The dimensions were obtained from Ref. [8]. A basic
schematic of the aircraft is shown in Figure 1.

Table 1. Aircraft Weights and Dimensions

Parameter Land Operation Water Operation

Maximum Ramp Weight (lb) 44,000 38,000
Maximum Takeoff Weight (lb) 43,850 37,850
Maximum Landing Weight (lb) 37,000 37,000
Maximum Zero Fuel Weight (lb) 43,000 43,000
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

Maximum Touchdown Weight for Water Pickup (lb) --- 36,200

Maximum Liftoff Weight after Water Pickup (lb) --- 47,000
Wing Area (ft2) 1,080
Wingspan (ft) 93.9
Wing Aspect Ratio 8.2
Overall Length 65.0

Figure 1. Schematic of CL-415 [9]

B. Flight Phases
The analysis was confined to airborne phases between takeoff and landing, be it on land or water. Ferry and
maintenance/training missions consisted of only one phase, incorporated in cruise 2. However, firefighting missions
were divided in to ten distinct phases, shown schematically in Figure 2, while the time history of an actual flight with
two fills and two drops is presented in Figure 3. Flights were defined as takeoff to landing, be it on land or on water.

Figure 2. Schematic of Flight Phases for Firefighting Missions

2
 8000 1600
Altitude
7000 1400
Water Load
6000 1200

MSL Altitude (ft)

Water Load (gal)

5000 1000

4000 800

3000 600

2000 400

1000 200

0 0
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

2600 3000 3400 3800 4200 4600 5000

Time (s)

Figure 3. Actual Time History for a Flight with Two Fills and Two Drops

In analyzing data from other air tankers, buffers were built between flight phases to avoid their overlapping in the
transition process. This resulted in a small part of the data being omitted from the analysis. In the present case, this
could not be done without leaving out a large fraction of the data due to the large number of phases. For example, a
firefighting mission with 20 fills would result in the analysis of 162 individual phases. Using a one-minute buffer
between the phases could result in omitting 2.7 hours of flight data in one mission. Therefore, no buffers were built
in for transitioning from one flight phase into another. Consequently, in some cases, elements of one phase affected
the results of the neighboring phases

C. Fill and Drop Sequence

The aircraft were equipped with four internal water tanks with total capacity of 1,621 gallons. Water could be
dropped from four bay doors in salvo, trail, or be split into separate drops depending on targets and fire type. Drop
heights ranged 100-150 feet [10]. During the fill phase, the amount of water could exceed the above limit. However,
the additional load was ejected through overflow vents on the side and the fill probes while they were retracted.
A typical sequence of a fill followed by a drop and the time history of various parameters is shown in Figure 4.
The filling operation started with fill entry. In this phase, the aircraft were configured for water pick-up by lowering
the flaps and the fill probes. Generally, the fill probes were lowered long before the start of the fill entry so their
position could not be used to mark the start of the fill phase. The flaps were usually placed in the second detent and
the aircraft was slowed for contacting the water surface. The fill probes were located just aft of the step on the keel,
so the aircraft would contact the water shortly before the start of the fill.
During the fill phase, the aircraft skimmed the water surface, usually without settling in, and the tanks were filled
through the fill probes. While the maximum tank capacity was 1,621 gallons, at the end of the fill phase the recorded
water level could reach as high as 2,000 gallons, much of which was ejected through the overflow vents and the
retracting fill probes during the exit from the fill. The aircraft weight could reach 47,000 pounds at the end of the fill.
In this phase, the aircraft could be subjected to large high-frequency vertical load factors stemming from water impact
as shown in Figure 4(b). These were not “flight loads,” and therefore, could not be categorized as gust or maneuver
loads. It was not uncommon for the airspeed to drop slightly below the stall speed for the aircraft weight, as indicated
in Figure 4(c). Also, due to the resolution of the terrain elevation data, sometimes Above Ground Level (AGL) altitude
became negative during this phase. This can be seen in Figure 4(d).
The next phase was the fill exit when the aircraft gained altitude while ejecting excess water. At the end of the
fill exit phase, the water level did not exceed 1,600 gallons. Usually, the flaps were raised to the first detent or were
retracted completely.
Just prior to entering the drop zone, the aircraft were configured for the drop phase. Again, flaps were lowered
to the second detent and altitude was reduced for entry into the fire zone. This was followed by the drop phase when
water left the aircraft. The end of the drop phase could not be determined solely based on the bay door positions. The
bay doors remained open for some time after the water load had been depleted. Therefore, the end of the drop phase

3
 had to be identified based on the water level in the tanks. Consistent with previous observations, the drop phase was
accompanied with an increase in the vertical load factor as shown in Figure 4(b).
The final phase in the process was drop exit, which was the egress from the fire zone. Flaps were raised from the
second detent into the first or were retracted completely. The variations in the vertical load factor during this phase
were comparable with those of the drop phase, although with the aircraft being at a much lower weight.
Depending on the proximity of the water source to the fire zone, the fill and drop sequence could take a very short
time. In the example shown here, the beginning of the fill entry and the end of the drop exit were only 45 seconds
apart.
Drop phases could be identified easily from the combination of the water level and the bay door signal. However,
fill phase required a more complex algorithm. At times, the crew would cycle the bay doors in flight. Invariably, this
would result in a jump in the recorded water level comparable to an actual fill. Therefore, special algorithms were
needed to bypass these false fills. Also, most of these occurred at indicated airspeeds above the bay door limit of 129
KIAS.
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

a. Definition of Phases b. Time History of Fill and Drop

c. Time History of Airspeed d. Time History of AGL Altitude

Figure 4. A Typical Fill and Drop Sequence

D. Recorded and Processed Data

All aircraft were equipped with IONode DFDRs supported by Latitude Technologies Corporation. Their stored
data was processed by Latitude software into a uniform 32-Hz format. The recorders were mounted very close to the
aircraft center of gravity, next to the water tanks.
The data used for this analysis is summarized in Table 2. Not all the data could be used for analysis for several
reasons including files that did not contain flights and where the quality of the data was questionable. The yield rate
of over 95% was higher than that seen from other aircraft.

4
 Table 2. Number of Files Used for Analysis

Files with Flight Time

Aircraft Date Range Useful Files
Flights (hr)
1 01-22-2015 to 12-13-2019 630 530 1,323.9
2 03-11-2016 to 11-20-2019 509 465 1,101.8
3 02-18-2016 to 11-20-2019 498 464 1,191.4
4 03-17-2016 to 12-17-2019 446 414 1,095.8
Total 2,083 1,873 4,713

Earlier efforts showed strong correlation between flight loads spectra and AGL altitude for low-level operations.
Data collected prior to 2018 did not contain terrain elevation needed for determining the AGL altitudes. The GPS
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

coordinates were used in the National Elevation Dataset (NED) [11] to obtain terrain elevation when it was not a
recorded parameter. Each set of coordinates had to be sent to the dataset via Internet, which returned the corresponding
ground elevation. Data transmissions rates made it impractical to perform this operation for every line of data.
Therefore, ground elevation was found once per second (i.e. for every 32 lines) and was interpolated in between.
Earlier work proved that the 32-Hz collected vertical load factors could contain accelerations stemming from
local or structural vibrations. This could lead to unrealistically high frequency of occurrence of the vertical gust load
factors [12, 13]. Therefore, an eighth-order low-pass Butterworth filter with an 8-Hz cutoff frequency was used to
remove the higher frequencies.
Flap deflections were recorded in degrees. Per Ref. [7], flaps could be set at 0, 10, 15, or 25 degrees with
corresponding airspeed and vertical load factor limitations. In practice, the recorded values did not match these
settings exactly. Therefore, the recorded signals were converted into discrete detents based on the limits shown in
Table 3.

Table 3. Flap Deflections in Degrees and Detents

Flaps per Recorded Values

Detents
Handbook (deg) (deg)
0 0 < Deflection < 7 Retracted
10 7 < Deflection < 13 1
15 13 < Deflection < 19 2
25 19 < Deflection 3

Frequency of occurrence of flight loads was determined from the method of peak-between-means of Ref. [14]
where only one extremum value is counted between every two crossings of the mean value. The “mean” value was
defined as any load within the dead band of ±0.05 g. The same dead band values were used for gusts and maneuvers
load factors. The values outside of the dead band that lasted two seconds or longer were attributed to maneuvers.
Other occurrences were assumed to be due atmospheric disturbances, or gusts. The rationale behind this two-second
rule is described in detail in Ref. [15]. This method was consistent with past practices and allowed direct comparison
of the results with previous work.

III. Results and Discussion

Flight loads spectra were developed for gust and maneuver load factors and were normalized per 1000 hours and
per nautical mile. In the interest of brevity, only the former results are discussed in this article. Furthermore, since
the cumulative occurrences of the flight loads correlated better with AGL altitudes, only these results are shown here.

A. Cruise Phases
Total distance and duration associated with various phases of firefighting missions, as well as those of ferry and
maintenance/training flights are summarized in Table 4. According to these results, the longest flight durations were

5
 those of cruise heavy and cruise light phases, followed by ferry missions. However, the former two flight phases were
punctuated by repeated fills and drops.

Table 4. Summary of Durations and Distances for all Flight Phases

Duration
Phase Duration (hr) Distance (nm)
(min)
Cruise 1 35,009 583.5 98,434
Fill Entry 2,471 41.2 3,944
Fill 2,960 49.3 3,817
Fill Exit 2,470 41.2 3,416
Cruise Heavy 91,825 1,530 229,730
Drop Entry 2,472 41.2 3,947
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

Drop 809 13.5 1,696

Drop Exit 1,307 21.8 2,814
Cruise Light 57,247 954 155,806
Cruise 2 28,810 480 80,203
Ferry 50,423 840 146,457
Maint./Training 4,893 82 12,235

The cumulative occurrences of incremental gust and maneuver load factors for cruise 1, for various AGL altitude
bands, are shown in Figures 5. The amount of data from the highest two altitude bands was extremely limited, resulting
in much scatter in the results. However, the results from the other altitude bands showed the dependence of the
frequency of occurrence on altitude. This can be seen more clearly in the maneuver loads spectra. The magnitudes
and the frequencies of occurrence were consistent with past observations. Cruise 1 phases showed the largest
difference in the frequency of occurrence of positive and negative incremental maneuver load factors at the lowest
altitudes compared with other cruise phases.
Similar results for cruise 2 flight phases are shown in Figure 6. However, in this case, the dependence of the gust
loads on altitude is somewhat clearer than in the case of cruise 1 phases with a slightly higher frequency of occurrence.
The maneuver loads spectra showed similar trends with slightly lower frequencies of occurrence but similar
magnitudes.
The cumulative occurrences of incremental gust and maneuver load factors for cruise heavy phases are shown in
Figure 7. These phases were flown almost exclusively below 4,500 feet AGL with a few cases above this altitude.
This is evident in Figure 7 where the results are concentrated in these altitude bands. The frequency of occurrence of
the gust loads was lower than those of the cruise 1 and cruise 2 phases, owing to the higher weights and slower flight
speeds. The maneuver load factors were also slightly smaller than those of other cruise phases.

a. Gust Loads b. Maneuver Loads

Figure 5. Cumulative Occurrence of Gust and Maneuver Loads – Cruise 1

6
 a. Gust Loads b. Maneuver Loads
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

Figure 6. Cumulative Occurrence of Gust and Maneuver Loads – Cruise 2

a. Gust Loads b. Maneuver Loads

Figure 7. Cumulative Occurrence of Gust and Maneuver Loads – Cruise Heavy

The results from cruise heavy phases were in contrast with those of cruise light phases where incremental gust
and maneuver load factors were larger and more frequent, as shown in Figure 8. For example, incremental gust load
factor of +1.0 g and larger were more frequent by an order of magnitude for cruise light phases. The same was true
to incremental maneuver load factors. In both cases, the effect of AGL altitude on maneuver load factors was more
pronounced than on gust load factors.

a. Gust Loads b. Maneuver Loads

Figure 8. Cumulative Occurrence of Gust and Maneuver Loads – Cruise Light

7
 The results for ferry flights are shown in Figure 9. These missions were defined as those that started and ended
in different places and were flown over a wide range of altitudes. Therefore, the correlation of the frequency of
occurrence of the load factors with AGL altitudes is more prominent in their cumulative occurrence of the incremental
gust load factors. These missions did not entail as much maneuvering as firefighting flight phases did. Consequently,
their incremental maneuver load factors spanned over a much narrower range than those of other cruise phases
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

a. Gust Loads b. Maneuver Loads

Figure 9. Cumulative Occurrence of Gust and Maneuver Loads – Ferry

The cumulative occurrences of incremental gust and maneuver load factors for the maintenance/training flights
are shown in Figure 10. These flights were very similar to ferry missions, except they started and ended at the same
place and were conducted at much lower altitudes. Therefore, they amounted to very limited volume of data and were
limited to AGL altitude bands below 9,500 feet. In the lower three altitude bands, good correlation can be seen with
the gust loads results from ferry flights. At higher altitude bands, scarcity of the recorded data rendered the results
unreliable.

a. Gust Loads b. Maneuver Loads

Figure 10. Cumulative Occurrence of Gust and Maneuver Loads – Maintenance/Training

B. Water Loading Phases

The three phases of fill entry, fill, and fill exit were almost exclusively flown in the lowest altitude band. Fill
entry and fill exit phases were limited to a maximum duration of 10 seconds, and therefore, their total times and
distances were very similar. Also, as touchdown and liftoff points of these phases could not be determined with
certainty, the load factors associated with these phases include some from water impact.
Cumulative occurrences of incremental gust and maneuver load factor for fill entry are shown in Figure 11. The
dominance of the positive gust load factors is an indication of the influence of water impact loads during this phase.
Also, while the maneuver load factor spectra are given in this figure, their reliability can be legitimately questioned
due to the same factor.

8
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

a. Gust Loads b. Maneuver Loads

Figure 11. Cumulative Occurrence of Gust and Maneuver Loads – Fill Entry

Figure 12 shows the cumulative occurrence of

incremental vertical load factor during fill phases. In
this case, the load factors were almost entirely due to
contact with water, as opposed to “flight” loads.
Therefore, their separation into gust and maneuver loads
would not be meaningful, so the results in this figure are
based on all load factors encountered during the fill
phase. The magnitudes and the frequencies compare
well with those of fill entry phase and both are much
larger than those of the cruise phases. However, in this
case, the load factors are more akin to landing impact
loads

The results from the fill exit are shown in Figure Figure 12. Cumulative Occurrence of Vertical
13. Cumulative occurrences of the incremental gust Loads – Fill
load factors bear some resemblance to those of fill entry,
although with lower frequencies and magnitudes. It is likely that the lower magnitudes are due to the higher wing
loading during this phase. Figure 13(b) is given here for the sake of completeness. Obviously, no conclusion can be
drawn from this figure regarding cumulative occurrences of the incremental maneuver loads related to fill exit phases.

a. Gust Loads b. Maneuver Loads

Figure 13. Cumulative Occurrence of Gust and Maneuver Loads – Fill Exit

9
 C. Water Drop Phases
Unlike the fill cases, these phases could be identified accurately from a combination of the bay door signal, water
tank level, and flap states. It is clear from this data that the majority of these phases were flown below 500 feet AGL.
The inclusion of the data from other than the lowest altitude band was only due to uncertainties in the terrain elevation.
It should also be noted that the scarcity of recorded data in the third altitude band rendered the related results somewhat
questionable.
The cumulative occurrences of the incremental gust and maneuver load factors for drop entry are shown in Figure
14. The gust load factors were found to be much larger and more frequent for this phase than those presented earlier.
At an incremental value of +1.0 g, the cumulative occurrences were one order of magnitude more frequent than those
of fill entry and two orders of magnitude larger than those of cruise phases. This proved to be true for both gust and
maneuver load factors.
Similar results were obtained from the drop phases, shown in Figure 15, but with significant magnification in
magnitudes and frequency of occurrence. Incremental gust load factors as large as +2.5 g were detected, as evident
from this figure. At incremental gust load factor of +1.0 g, the cumulative occurrence for the lowest altitude band was
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

more than three times the order of magnitude of the same during cruise 1 phase. The same was true for incremental
maneuver load factors. Both sets of results were quite consistent with those from other air tankers during the drop
phase. Also, the positive maneuver load factors shown in this figure were mostly driven by a spike in this parameter
associated with the drop phases. While this is seen in data from all types of air tankers, it is unclear whether the
increased load factors are due to crew action or some external cause such as aerodynamics or changes in the location
of the aircraft center of gravity.

a. Gust Loads b. Maneuver Loads

Figure 14. Cumulative Occurrence of Gust and Maneuver Loads – Drop Entry

a. Gust Loads b. Maneuver Loads

Figure 15. Cumulative Occurrence of Gust and Maneuver Loads – Drop

10
 Finally, the results from the drop exit phases are shown in Figure 16. The frequency of occurrence of the
incremental gust load factors were similar to those of the drop entry phase, except for more frequent and larger negative
values. Again, the frequency of occurrence was about two orders of magnitude larger than those of cruise phases.
The same was true for the maneuver load factors. The maneuver load factors during the drop exit phases were second
in magnitude to those of the drop phases. This behavior was also in agreement with previous observations of other
air tankers.
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

a. Gust Loads b. Maneuver Loads

Figure 16. Cumulative Occurrence of Gust and Maneuver Loads – Drop Exit

D. Comparison with Other Aircraft

In this section flight loads spectra are compared among flight phases and with other operations. In the interest of
brevity, only cumulative occurrences per 1000 hours are shown and discussed. In addition, in the case of CL-415,
only airborne phases are presented to eliminate any influence from water impact loads associated with fill entry, fill,
and fill exit phases.
Cumulative occurrences of incremental vertical gust load factors for various flight phases and missions are shown
in Figure 17. This figure shows the results for all altitude bands combined. Also, for clarity, the results from the
phases related to drops are shown separately. In the first part of this figure, results are compared from different cruise
phases. Ferry and maintenance/training missions are included in this part due to their similarity with cruise phases.
This figure shows clearly that the highest frequency of occurrence of the vertical gust load factors were in conjunction
with cruise light phases, in all likelihood due to the lower wing loadings and slightly higher flight speeds. The lowest
frequencies occurred during cruise heavy phases and ferry missions. This could be attributed to the higher wing
loading of cruise heavy phases and the higher overall altitudes of the ferry missions.

a. Cruise Phases b. Drop Phases

Figure 17. Cumulative Occurrences of Incremental Vertical Gust Load Factor by Flight Phase

11
 In the second part of Figure 17, cumulative occurrences of incremental vertical gust load factors are compared
among drop entry, drop, and drop exit phases. This figure also shows the results from ferry missions for reference. It
is evident from this figure that the drop phases involved the largest load factors, with two to three orders of magnitude
in frequency relative to ferry missions. This was consistent with previous observations [1, 2, 12, 16]. It was also
noted that the positive load factors were more frequent than the negative load factors, whereas they were equally
represented in cruise phases.
Cumulative occurrences of incremental maneuver load factors are compared among flight phases in Figure 18.
Again, for clarity, flight phases surrounding the drop are shown separately in comparison with ferry missions for
reference. Much like in the previous case, cruise light phases showed the highest frequency of occurrence of the
maneuver load factors, while ferry missions had the fewest. The other cruise phases and the maintenance/training
missions occurred with almost equal magnitude and frequency. However, maneuver load factors, shown in the second
part of this figure, occurred with significantly large magnitude and frequency during the drop, followed by the drop
exit phases. At the incremental load factor of +1.0 g, maneuver load factors from the drop phases were more than
four orders of magnitude more frequent than those of ferry flights. The higher frequencies and magnitudes were
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

consistent with those of other airtankers, although they were both quite larger in this case. However, it should be
noted that these phases constituted a very small fraction of the total flight time.

a. Cruise Phases b. Drop Phases

Figure 18. Cumulative Occurrences of Incremental Vertical Maneuver Load Factor by Flight Phase

The CL-415 is a completely different design from other airtankers or civilian operations that were examined
earlier. Therefore, it was difficult to find a single “similar” aircraft to which its flight loads spectra could be compared.
The aircraft could be considered as a next-generation airtanker but was substantially lighter than BAe-146 and RJ-85
[12] with flight profiles that were vastly different. Agricultural aircraft that were addressed in Ref. [16] were flown
with flight profiles somewhat similar to that of CL-415, with many turns and much low-level maneuvering, but were
significantly lighter and were subject to much more gradual changes in aircraft weight. Single-Engine Airtankers
(SEATs) [16] were subject to large in-flight changes of weight but performed only one or two drops per mission.
Finally, the ASM/lead aircraft [4] had flight profiles most similar to that of CL-415 without carrying a water load.
Therefore, in the next figures, flight loads spectra from CL-415 are compared with those of all of the above aircraft.
It should be noted that on the agricultural, the SEATs, and the ASM/lead aircraft the data was collected at 8 Hz. In
Ref. [13] the authors showed that this could lead to slightly lower frequencies of occurrence of the gust load factor,
but maneuver load frequencies remained unaffected.
Cumulative occurrences of incremental vertical gust load factors are compared in Figure 19. Except for the
agricultural operations, all data was obtained from aerial firefighting missions. Also, to prevent water impact loads
from affecting the CL-415 spectra, the results from fill entry, fill, and fill exit phases were excluded for this
comparison. It is obvious from this figure that the gust loads had the same frequency and magnitude as those of the
ASM/lead aircraft that had very similar flight profiles and were flown in the same environment. The two aircraft also
had similar wing loadings and were flown in the same range of altitude for long periods. On the other hand, the lowest
frequencies were associated with the BAe-146/RJ-85 and PZL-M18 used for agricultural operations. The former
could be attributed to the much larger wing loading and the fact these aircraft were in cruise altitudes over most of

12
 their flights. The lower cumulative occurrences of the incremental gust load factors for the SEATs and the agricultural
operations were puzzling.
Likewise, Figure 20 shows the cumulative occurrences of incremental maneuver load factors for the five
operations. The frequency of occurrence agreed well with those of ASM/lead aircraft, especially at higher values.
Again, the lowest frequencies were from the BAe-146/RJ-85 that typically spent a small part of their flight profiles in
the fire zone and were flown in cruise phases most of the time with little need for maneuvering.
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

Figure 19. Cumulative Occurrences of Incremental Figure 20. Cumulative Occurrences of Incremental
Vertical Gust Load Factor Compared with Other Vertical Maneuver Load Factor Compared with
Operations Other Operations

IV. Conclusions
Operational flight data collected from four CL-415 airframes were used for flight loads analysis. Missions were
of three categories; aerial firefighting, ferry, and maintenance/training flights. Firefighting flights were further divided
into ten individual phases. Results were extracted and reported for each flight phase and for ground-air-ground cycles
where appropriate. The data amounted to just over 4,700 hours of flight operations.
Vertical accelerations were separated into those due to gusts and maneuvers. In each category, the cumulative
occurrences of the incremental load factors were established for various AGL altitude bands. For the four cruise
phases, the ferry, and the maintenance/training missions, which were flown over a wide range of altitudes, flight loads
spectra were also developed for MSL altitude bands. The results were normalized per 1000 hours.
The longest cumulative flight durations were those of cruise heavy and cruise light phases, punctuated by repeated
fills and drops. The third longest phase belonged to ferry missions, which also covered the widest range of altitudes.
In general, firefighting cruise phases were flown at relatively low AGL altitudes. Most cruise 1 and cruise 2 phases
remained under 9,500 feet AGL while the vast majority of cruise heavy and cruise light phases took place below 4,500
feet AGL.
The cumulative occurrences of the incremental gust and maneuver load factors from cruise 1 and cruise 2 showed
a clear correlation with AGL altitude bands. The same trends could not be seen clearly when the results were placed
in MSL altitude bands. This was consistent with previous observations.
Cruise heavy phases showed a slightly lower cumulative occurrences of incremental load factors than cruise 1 and
cruise 2 phases. Cruise light phases were flown at somewhat lower altitudes than cruise heavy phases and contained
larger and more frequent incremental load factors than the latter. These phases were also flown at slightly higher
airspeeds.
Cumulative occurrences of incremental gust and maneuver load factors were presented for fill entry, fill, and fill
exit phases. However, since the associated load factors included those due to contact with the water, they could not
be strictly called flight loads. This was especially true for the fill phase where almost all vertical accelerations were
due to water impact. These accelerations were shown to be significant.
Cumulative occurrences of gust and maneuver load factors for drop entry, drop, and drop exit phases were
developed. It was shown that the largest vertical maneuver load factors were in conjunction with the drop phases,
followed by those of drop exit and drop entry. The frequency of occurrence of the vertical load factors in these phases
was between one and three orders of magnitude larger than those of cruise phases. These results were consistent with
previous observations of firefighting missions.

13
 Cumulative occurrences of incremental gust and maneuver load factors were compared among different flight
phases. For this purpose, the results from all altitude bands were combined. Among the cruise phases, the most
frequent vertical gust load factors were observed to be associated with the cruise light phases that were flown at
relatively low weights and somewhat higher airspeeds. Cumulative gust load factors from the drop phases were shown
to be two to three orders of magnitude more frequent than those of cruise phases, followed by drop exit and drop entry
phases. The same trends were also observed in the cumulative incremental vertical maneuver load factors.
Flight loads spectra for the airborne phases were also compared with those of firefighting missions from BAe-
146/RJ-85, SEATS, and ASM/lead aircraft, and agricultural spraying. Each of these operations was deemed to have
some similarities with CL-415 firefighting missions. All flight phases were combined for these comparisons, but the
gust and the maneuver load factors were kept separate. In addition, fill entry, fill, and fill exit phases were left out to
prevent inclusion of water impact loads.
Vertical gust load factors were shown to have magnitudes and frequencies very similar to those of ASM/lead
aircraft. However, these loads were shown to be more frequent than those of heavy airtankers by two to three orders
of magnitude. The same trends were observed regarding maneuver load factors although the results from CL-415
Downloaded by WICHITA STATE UNIVERSITY on July 5, 2022 | http://arc.aiaa.org | DOI: 10.2514/6.2022-4015

were slightly less frequent at lower load factors.

Acknowledgement
This effort was funded by the United States Forest Service through the Grant 17-G-004 administered by the Federal
Aviation Administration. The authors wish to recognize the support and the guidance provided by Mr. David
Rathfelder from Los Angeles Aircraft Certification Office and Ms. Heather Castillo from United States Forest Service
in conducting this work. The FAA project monitor was Dr. Sohrob Mottaghi. The assistance of our research assistant,
Mr. Mason Bivens, was also invaluable in parts of this project.

References
[1] Rokhsaz, K., Kliment, L. K., Nelson, J., and Newcomb, J., “Operational Assessment of Heavy Air Tankers,” AIAA Journal
of Aircraft, Vol. 51, No. 1, January-February 2014, pp. 62-77, DOI: 10.2514/1.C031922.
[2] Rokhsaz, K., Kliment, L. K., and Bramlette, R. B., “Usage and Maneuver Loads Monitoring of Heavy Air Tankers,” FAA
Report DOT/FAA/AR-11/7, March 2011.
[3] Kliment, L.K., Rokhsaz, K., Nelson, J., Terning, B., and Weinstein, E.M., “Usage and Flight Loads Analysis of King Airs in
Aerial Firefighting Missions,” AIAA Journal of Aircraft, Vol. 52, No. 3, May-June 2015, pp. 910-916, DOI:
10.2514/1.C032877.
[4] Kliment, L. K., Rokhsaz, K., and Menon, A. N., “Further Airframe Usage and Operational Loads Monitoring of ASM/Lead
Aircraft,” FAA Report DOT/FAA/TC-17/40, November 2017, DOI: 10.13140/RG.2.2.12345.42087.
[5] Large Airtanker Modernization Strategy, USDA Forest Service, Final Version Released February 10, 2012, January 17,
2012.
[6] Federal Aviation Administration, Type Certificate Data Sheet A14EA, Revision 11, October 1, 2016.
[7] Canadair 415 Model CL-215-6B11, Airplane Flight Manual, Product Support Publication No. 491, Bombardier Aerospace,
Montreal, Quebec, 1999.
[8] Jane’s All the World’s Aircraft, Editors M. Lambert and K. Munson, 1994-1995 Issue, Jane’s Information Group, Inc.,
Alexandria, VA, 1994.
[9] NWCG Airtanker Base Directory, National Wildfire Coordinating Group, PMS 507, August 2020.
[10] CL-415 Amphibious Water Scooping Aircraft User Guide, United States Forest Service Fire and Aviation Management,
AWSA User Guide V4 29 June 2018.
[11] Anonymous, “National Elevation Dataset,” https://gishubdata.nd.gov/dataset/usgs-national-elevation-dataset-ned, last
accessed July 2019.
[12] Rokhsaz, K., Kliment, L. K., and Rollins, E., “Flight Loads and Airframe Usage Analysis of Next-Generation Airtankers –
Phase I,” FAA Final Report (under review), August 2018.
[13] Rokhsaz, K. and Kliment, L. K., “Effect of Data Filtering on Flight Loads Spectra,” AIAA-2019-3367, AIAA Aviation and
Aeronautics Forum and Exposition, Dallas, TX, June 17-21, 2019, doi.org/10.2514/6.2019-3367.
[14] “Standard Practices for Cycle Counting in Fatigue Analysis,” ASTM E 1049-85 (Reapproved 2005).
[15] Rustenburg, J. W., Skinn, D. A., and Tipps, D. O., “An Evaluation of Methods to Separate Maneuver and Gust Load Factors
from Measured Acceleration Time Histories,” FAA Report DOT/FAA/AR-99/14, April 1999.
[16] Rokhsaz, K. and Kliment, L. K., “Preliminary Comparison of Agricultural and Single-Engine Air Tanker Operational Loads,”
DOT/FAA/TC-16/9, Contract Monitor: Dr. Edward Weinstein, Final Report to Federal Aviation Administration, December
2016.

14