J Vis

https://doi.org/10.1007/s12650-024-01039-0

R E G UL A R P A P E R

Suraj Bansal • Philippe Lavoie

Smoke visualization of the effect of freestream

turbulence on a laminar separation bubble
over an airfoil at low Reynolds numbers

Received: 3 October 2024 / Revised: 3 December 2024 / Accepted: 9 December 2024

Ó The Visualization Society of Japan 2025

Abstract The two-dimensional, time-resolved structure of laminar separation bubbles over an SD7003
airfoil under the influence of elevated background turbulence levels is presented using smoke visualization.
The separation bubbles are visualized for angles of attack a ¼ 4 and 8 , Reynolds numbers Re ¼ 60; 000
and 100,000, and turbulence intensities Tu ¼ 0:02% and 0:24%. High-speed images captured the formation
of shear layer roll-up vortices in the aft portion of the separation bubble for Tu ¼ 0:02%, which are
compared with the Tu ¼ 0:24% case. On increasing the background turbulence (Tu ¼ 0:24%), the bubble is
eliminated for a ¼ 4 , while its height and chordwise extent are considerably decreased for a ¼ 8 . The
images show that the large-scale turbulent structures in the freestream interact with the shear layer, causing a
greater variance in the size and shedding period of the roll-up vortices. Increased dissipation rates cause
greater diffusion and distort the roll-up vortices. Increasing the Reynolds number for Tu ¼ 0:24% does not
affect the size and shedding wavelength of the roll-up vortices.
Graphical abstract

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s12650-
024-01039-0.

S. Bansal  P. Lavoie (&)

Institute for Aerospace Studies, University of Toronto, 4925 Dufferin Street, Toronto, ON M3H 5T6, Canada
E-mail: lavoie@utias.utoronto.ca
S. Bansal
E-mail: suraj.bansal@mail.utoronto.ca
 S. Bansal, P. Lavoie

Keywords Airfoil  Smoke visualization  Freestream turbulence  Laminar separation bubbles 

Vortex shedding

1 Introduction

At low Reynolds numbers (Re\106 ), boundary layers can readily separate from the airfoil upper surface
due to the presence of a strong adverse pressure gradient. The separated shear layers may stay separated or,
under certain circumstances, can reattach to form laminar separation bubbles (LSBs) (Arena and Mueller
1980; Carmichael 1981; Mueller and Batill 1982; Lissaman 1983; Bastedo Jr. and Mueller 1986; Mueller
and DeLaurier 2003). The time-averaged topology of a separation bubble is depicted in Fig. 1. The sepa-
rated shear layer rapidly transitions to turbulence, inducing a mean-flow reattachment. The region of
recirculating fluid between the separated shear layer and the airfoil upper surface is defined as a laminar
separation bubble in a time-averaged sense. The presence of an LSB has an adverse effect on the aero-
dynamic performance of an airfoil (Carmichael 1981; Mueller and DeLaurier 2003); therefore, much
attention has been given to understanding its behavior. For instance, the influence of varying the airfoil
geometric and freestream flow parameters such as angle of attack a, Re, and freestream turbulence intensity
Tu on the time-averaged LSB has been investigated (Arena and Mueller 1980; Bastedo Jr. and Mueller
1986; Burgmann et al 2008; Burgmann and Schroder 2008; Zilli et al 2017; Kirk and Yarusevych 2017;
Istvan and Yarusevych 2018).
The instantaneous flow topology of a separation bubble differs notably from the time-averaged one. The
amplification of small-amplitude disturbances in the separated shear layer can cause the shear layer to roll
into coherent spanwise vortices, typically shed in the aft portion of the bubble (Wilson and Pauley 1998;
Burgmann and Schroder 2008; Lengani et al 2014; Lambert and Yarusevych 2017; Kirk and Yarusevych
2017; Istvan and Yarusevych 2018). The shear layer roll-up is found to trigger when the disturbances in the
shear layer grow to a certain threshold level (Burgmann and Schroder 2008; Marxen et al 2013; Kirk and
Yarusevych 2017). Once their formation occurs, these spanwise oriented vortex filaments develop undu-
lations during the later stages of the transition process and subsequently break into smaller-scale three-
dimensional turbulent structures (Jones et al 2008; Marxen et al 2013; Kurelek et al 2016) near the mean
reattachment location (Kirk and Yarusevych 2017). These coherent vortices play an important role in the
LSB dynamics, where they facilitate the entrainment of high momentum fluid from the outer flow and
consequently assist in the separated shear layer reattachment.
While the effects of elevated Tu on the mean and instantaneous behavior of LSBs have been previously
considered using surface pressure, hot-wire anemometry, and particle image velocimetry measure-
ments (Hain et al 2009; Olson et al 2013; Zilli et al 2017; Simoni et al 2017; Istvan and Yarusevych 2018;
Dellacasagrande et al 2020; Kumar and Sarkar 2024; Jaroslawski et al 2023; Dellacasagrande et al 2024;
Kumar et al 2022), the interaction of the turbulent structures in the freestream with the shear layer roll-up
process has not been visualized before. This manuscript presents smoke visualization images that capture
the instantaneous behavior of laminar separation bubbles formed over an SD7003 airfoil under an elevated
freestream turbulence level (Tu ¼ 0:24%) and compares them with the baseline turbulence cases
(Tu ¼ 0:02%).

Fig. 1 Time-averaged topology of a laminar separation bubble

Smoke Visualization of the Effect of Freestream Turbulence...

Fig. 2 A schematic diagram of the test section representing the experimental setup used in the present investigation

2 Experimental setup

All experiments were performed in the low-speed recirculating wind tunnel at the University of Toronto
Institute for Aerospace Studies (UTIAS). The wind tunnel has a hexagonal test section that is 1.2 m wide,
0.8 m tall, and 5 m long. Tests were conducted with mean velocities between 3 and 5 m/s, corresponding to
Reynolds numbers of Re ¼ 60; 000 and Re ¼ 100; 000; respectively.
The airfoil selected for the present study is the SD7003 profile with a chord length (c) of 305 mm and a
span (b) of 889 mm, resulting in an aspect ratio (AR) of 2.9. The airfoil assembly was placed between two
endplates to reduce three-dimensionality in the flow field (see Fig. 2). More details about the airfoil
apparatus, verification of the flow two-dimensionality, and experimental setup validation are provided
in Bansal and Lavoie (2022).

2.1 Turbulence generation

The higher turbulence intensity was generated using a wire-mesh grid (denoted by W18M6) with a mesh
length M ¼ 6 mm and solidity r ¼ 18%, where r ¼ d=M ð2  d=M Þ and d is the wire diameter. The grid
was placed 1.6 m (5.1c or 250M) upstream of the airfoil leading edge. Thus, the turbulence generated by the
grid is assumed to be nominally homogenous and isotropic around the airfoil. The turbulence was measured
using a hot-wire probe operated by a Dantec Dynamics Streamline Pro constant temperature anemometer.
The sampling frequency for data acquisition was 50,000 Hz. The probe used a 5 lm copper-plated tungsten
wire with a sensing length of 1 mm. The streamwise component u (u ¼ u þ u0 , where u is the time-averaged
and u0 is the fluctuating component) of the freestream velocity was measured with the hot-wire in the empty
test section at the airfoil leading edge location. The corresponding power spectral density (Suu ) plots for u0
are shown in Fig. 3. pffiffiffiffiffiffi
The freestream turbulence intensity, defined as Tu  u02 =u, was calculated by integrating Suu between
fc=u ¼ 0:09 and fc=u ¼ 9 for the baseline turbulence case (no turbulence grid). Here, f is the frequency. For
the higher turbulence case, Suu was integrated between fc=u values of 0.09 and 56. To determine the lower
integration limit in each case, the time scale sts associated with the eddy sizes comparable to the smallest test
section dimension (hts ¼ 0:8 m) was estimated using sts ¼ hts =u. Subsequently, the lower integration limit
was conservatively chosen based on the estimated time scales of these eddies. The higher integration limit
corresponded to the noise floor in Suu (see Fig. 3). Rs
The integral length scale Lu is calculated using Lu ¼ u 0 0 Ruu ds, where Ruu is the autocorrelation
0
function of u , s is the time lag and s0 corresponds to the first zero crossing of Ruu . Here, Taylor’s hypothesis
of frozen turbulence is used while multiplying the integral of Ruu with u. An exponential function was fit to
pffiffiffiffiffiffi
the measured autocorrelation before integration (Fuchs et al 2022). The Reynolds number based on the
pffiffiffiffiffiffi length scale ReL is defined as u02 Lu =m and the microscale Reynolds number Rek is defined as
integral
u02 ku =m, where ku is the Taylor microscale. Table 1 provides the parameters characterizing the freestream
turbulence for both the baseline and higher Tu cases. Since Tu varies from 0.22% to 0.25% between
Re ¼ 60; 000 and 100,000, a mean value of 0.24% will be used to refer to the elevated turbulence cases.
 S. Bansal, P. Lavoie

Table 1 Turbulence characterization for the baseline and higher Tu cases obtained from hot-wire measurements made 250M
downstream from the grid. All length scales are expressed in mm. Here, Tu is the turbulence intensity, Lu is the integral length
scale, ku is the Taylor microscale, g is the Kolmogorov length scale,  is the turbulent dissipation rate, ReL is the Reynolds
number based on the integral length scale, and Rek is the microscale Reynolds number

Grid Re Tu (%) Lu ku g  (mm2 =s3 ) ReL Rek

No Grid 60,000 0.02 4621 166 30.5 0.005 214 7.7
W18M6 60,000 0.22 12.2 8.0 2.2 187 5.3 3.5
W18M6 100,000 0.25 9.7 5.9 1.4 1270 8.1 4.9

Fig. 3 Power spectral density plots for the streamwise velocity u (u ¼ u þ u0 ) showing the baseline (Tu ¼ 0:02%) and elevated
turbulence cases (0:22%  Tu  0:25%) for Re ¼ 60; 000 and 100,000. Hot-wire measurements were taken in an empty test
section at the airfoil leading edge location, 250M downstream from the turbulence grid

2.2 Smoke visualization

A smoke wire was vertically installed upstream of the airfoil to visualize the LSBs. A 30 AWG diameter
(dw ¼ 0:25 mm) Nichrome wire was used to achieve an optimal trade-off between reducing wire vortex
shedding and minimizing wire breakage. The Reynolds number Red based on dw is 54. Furthermore, the wire
was placed 0.7 m (2.25c or 2720dw ) upstream of the leading edge to further reduce the effect of the wire
vortex shedding on the global flowfield around the airfoil. In the spanwise direction (z), the wire was
installed at 2z=b ¼ 0:55 from the model center, where b is the model span, to avoid any potential inter-
ference from the strut at the model center (see Bansal and Lavoie 2022) or any junction vortices near the tip
region. Smoke was generated by evaporating propylene glycol beads on the nichrome wire using an
alternating current. A rheostat with a 110 V/60 Hz AC input was used to control the current passing through
the smoke wire. The flow was illuminated using four LED video lights (4000 Lux per LED video light)
placed on top of the glass ceiling panels of the test section. A Photron Fastcam SA5 high-speed camera
equipped with a 105 mm or 180 mm focal length lens (depending on the size of the bubbles) was used in the
present study, with the aperture set to the highest at f/2.8. The highest frame rate for which the LED video
lights could illuminate the camera field of view sufficiently was 3,000 frames per second (FPS). Further-
more, a frame rate of 3000 FPS was sufficient to capture the essential dynamics in forming the separation
bubbles and their interaction with the freestream turbulence. Therefore, smoke videos were acquired at
3,000 FPS.

3 Results

3.1 Baseline turbulence level ðTu ¼ 0:02%Þ characterization

Figure 4 shows the presence of a laminar separation bubble for Re ¼ 60; 000 and 100,000 at a ¼ 4 and
8 under the baseline turbulence level (Tu ¼ 0:02%). As expected, boundary layer separation is observed for
all cases as shown in Fig. 4. For a ¼ 4 and Re ¼ 60; 000, the boundary layer separation is estimated at
x=c  0:24. For this case, the separated shear layer shows signs of spatially periodic velocity fluctuations
Smoke Visualization of the Effect of Freestream Turbulence...

Fig. 4 Smoke visualization depicting laminar separation bubbles under the baseline turbulence level for a
a ¼ 4 ; Re ¼ 60; 000, b a ¼ 4 ; Re ¼ 100; 000, c a ¼ 8 ; Re ¼ 60; 000, and d a ¼ 8 ; Re ¼ 100; 000. The scale shown in
red represents the chordwise location (x/c) over the airfoil upper surface. The curved red line identifies the airfoil upper surface
at the plane where the smoke is introduced. Letters S, R, and T denote laminar boundary separation, shear layer roll-up, and
approximate turbulent reattachment points, respectively

Table 2 Comparison of the properties of a laminar separation bubble over an SD7003 airfoil for a ¼ 4 and Re ¼ 60; 000
obtained from different experimental facilities and computations (Ol et al 2005; Drela 1989; Hain et al 2009). The variable AR
denotes the model aspect ratio used in respective experiments

Dataset Method Tunnel AR Tu (%) xs =c xr =c hb =c

UTIAS Smoke vis Wind 2.9 0.02 0.24 0.63 0.029
AFRL TR-PIV Water 2.3–3 0.10 0.18 0.58 0.029
TU-BS 2005 TR-PIV Wind 2.0–2.9 0.10 0.30 0.62 0.028
IAR TR-PIV Tow-tank 3.6 – 0.33 0.63 0.027
UTIAS XFOIL N/A N/A 0.07 0.21 0.62 –
UTIAS Smoke vis Wind 2.9 0.24 N/A N/A N/A
TU-BS 2009 TR-PIV Water 1.3 0.28 0.35 0.53 0.021

near x=c  0:36, indicating the amplification of disturbances due to the adverse pressure gradient. Further
growth of these disturbances causes the shear layer to roll into coherent vortices, as seen at x=c  0:44.
These vortices undergo deformation and become more diffused at x=c  0:52, indicating the development of
spanwise undulations and the early phase of the breakdown of vortices during the later stages of transition.
Subsequently, the complete dissipation of the smoke filaments is observed at x=c  0:63; indicating a fully
turbulent flow. The complete breakdown of coherent vortices into three-dimensional structures occurs near
the mean reattachment point (Kirk and Yarusevych 2017); therefore, x=c  0:63 is assumed to be the mean
reattachment point. When Re is increased to 100,000 for a ¼ 4 , the separation point expectedly moves
downstream (x=c  0:27), while the roll-up location moves to x=c  0:36. Finally, the location of the
reattachment point also moves upstream to x=c  0:54: As a result, the chordwise extent of the LSB reduces
significantly. Furthermore, the bubble height has notably decreased.
For a ¼ 8 and Re ¼ 60; 000  100; 000, the separation point was outside the field of view and could not
be determined. For Re ¼ 60; 000, the roll-up location was observed to be x=c  0:12, with the reattachment
occurring at x=c  0:28. This upstream movement of the roll-up and reattachment points is expected since a
stronger adverse pressure gradient facilitates an earlier shear layer transition relative to the separation point,
 S. Bansal, P. Lavoie

Table 3 Comparison of the maximum height and wavelength of the roll-up vortices shed within the laminar separation bubble

a Re Tu ð%Þ hr /c kr =c Tu ð%Þ hr /c kr =c
4 60,000 0.02 0.029 0.078 0.24 N/A N/A
4 100,000 0.02 0.025 0.064 0.24 N/A N/A
8 60,000 0.02 0.028 0.036 0.24 0.023 0.043
8 100,000 0.02 – – 0.24 0.023 0.043

Fig. 5 Smoke visualization images for Tu ¼ 0:24% representing attached turbulent boundary layer for a a ¼ 4 ; Re ¼ 60; 000,
and b a ¼ 4 ; Re ¼ 100; 000. Laminar separation bubble are shown for c a ¼ 8 ; Re ¼ 60; 000, and d a ¼ 8 ; Re ¼ 100; 000

thereby causing a significant reduction in the bubble size. With the separation point also likely moving
upstream due to a stronger adverse pressure gradient, the overall bubble location shifts closer to the leading
edge.
The separation point xs =c, reattachment point xr =c, and maximum bubble height hb =c for a ¼ 4 and
Re ¼ 60; 000 from the present investigation are compared with other experimental findings, and
XFOIL (Drela 1989) using N ¼ 9 in Table 2. The three experimental datasets denoted by IAS, TU-BS
2005, and AFRL are from Ol et al (2005) and the dataset from Hain et al. (2009) with a higher Tu of 0.28%,
denoted as TU-BS 2009, is also included for comparison. The experimental datasets from the present study
(UTIAS) include relevant location estimates for both the baseline and elevated freestream turbulence cases;
the latter will be discussed in Sect. 3.2. These location estimates have a certain degree of uncertainty due to
cycle-to-cycle variations, especially in the shedding period of the coherent vortices and their roll-up
location. For instance, the variation observed in the roll-up location for a ¼ 4 and Re ¼ 60; 000 is 0.013c
based on the three shedding cycles captured in the corresponding smoke video. For a ¼ 4 and
Re ¼ 100; 000, the roll-up location varied from the mean position by 0.01c in the 16 shedding cycles that
were analyzed. Similarly, identifying the exact location of shear layer reattachment is uncertain due to the
highly unsteady nature of turbulence, where the maximum variation observed about the mean position is
0.02c for a ¼ 4 and Re ¼ 60; 000 and 0.027c for a ¼ 4 and Re ¼ 100; 000. Nevertheless, for the lower
turbulence levels (Tu  0:1%), the separation point, reattachment point, and maximum bubble height esti-
mates obtained from Fig. 4 are in good agreement with the studies referenced in Table 2.
Smoke Visualization of the Effect of Freestream Turbulence...

Fig. 6 Smoke visualization images for a ¼ 8 , Re ¼ 60; 000 , and Tu ¼ 0:24% depicting the influence of large-scale structures
in the turbulent freestream on the roll-up vortices. The time period of the vortex shedding shown here is approximately 0.006 s
( 180 Hz)

The maximum height (hr =c) of the roll-up vortices and their wavelengths (kr =c) are presented in Table 3.
Increasing Re for a given a causes a reduction in the size of the rollers and their wavelength. On the other
hand, increasing a for a given Re significantly reduces the wavelength and causes only a modest size
reduction.

3.2 Elevated turbulence level ðTu ¼ 0:24%Þ

As discussed earlier, the turbulence intensity for the elevated turbulence levels is estimated to be
Tu  0:24%. As shown in Table 1, the corresponding Lu ranges between 12.2 and 9.7 mm for
Re ¼ 60; 000  100; 000. The maximum turbulent boundary layer thickness on the airfoil approximated
using d ¼ 0:37c=Re1=5 is 12.5 mm for comparison. Thus, while the actual boundary layer thickness on the
airfoil will most likely differ, Lu can be assumed to be comparable to the boundary layer thickness on the
airfoil.
Figure 5 shows the smoke visualization images for Tu ¼ 0:24%. The smooth smoke streaklines in the
outer flow seen earlier for the baseline case in Fig. 4 are replaced by wavy streaklines expectedly, indicating
increased velocity fluctuations. The separation bubble was completely eliminated for Tu ¼ 0:24%, a ¼ 4 ,
and Re ¼ 60; 000  100; 000, where an attached turbulent boundary layer can be observed. Due to increased
velocity fluctuations at higher Tu ; the enhanced mixing between the boundary layer and outer flow increases
the boundary layer momentum, delaying flow separation. This momentum influx into the boundary layer
from the outer flow is likely further increased when the largest eddies in the turbulent freestream become
comparable to or larger than the boundary layer thickness. At smaller angles of attack, when the pressure
gradients are small enough, this higher momentum influx in the boundary layer can keep the boundary layer
attached.
The aforementioned elimination of the separation bubble is consistent with the SD7003 surface pressure
measurements from a previous investigation conducted in the same facility used in the present study (Zilli
 S. Bansal, P. Lavoie

et al 2017). Given that the turbulence intensity of 0.24% is a relatively low background turbulence, these
results highlight the sensitivity of laminar separation bubbles to increased perturbation levels.
Increasing Re to 100,000, while maintaining a similar turbulencep ffiffiffiffiffiffi
intensity, led to a greater amplification
of disturbances in the freestream, as shown in Fig. 5. Given that u02 is proportional to the freestream
velocity for a fixed Tu , increased levels of fluctuations in the streaklines at higher Re are expected. An
increased range of scales in the outer flow perturbations is also observed due to a higher Reynolds number of
the turbulence (see Table 1). This increased separation of scales can also be corroborated by comparing the
velocity spectra, Suu for Re ¼ 60; 000 and 100,000 in Fig. 3. Furthermore, the streaklines are more diffused
for Re ¼ 100; 000 due to a significantly higher dissipation rate , as shown in Table 1. The dramatically
higher dissipation rates at higher Re and for turbulence grids with high Tu levels make smoke visualization
of turbulent flow challenging since the smoke streaklines diffuse rapidly.
For a ¼ 8 and Re ¼ 60; 000, a significantly diminished separation bubble relative to the baseline
turbulence case is observed in Fig. 5. This separation bubble is smaller both in the streamwise extent and in
height. The higher initial disturbances in the shear layer lead to reaching the threshold disturbance levels
sooner, precipitating the transition process (Istvan and Yarusevych 2018). The early transition is correlated
to an early shear layer roll-up (Istvan and Yarusevych 2018) relative to the separation point, reducing the
size of the rollers (see Table 3) and consequently decreasing the maximum bubble height. Furthermore, the
wavelength of the shed vortices also increases, as shown in Table 3. This increase in the shedding wave-
length for higher freestream turbulence intensities is also observed in other experiments (Istvan and
Yarusevych 2018). Increasing Re for Tu ¼ 0:24% did not affect the vortex size or the wavelength, as evident
from Table 3.
Figure 6 shows the cycle-to-cycle variations in the roll-up vortices for a ¼ 8 , Re ¼ 60; 000 , and
Tu ¼ 0:24%. The large-scale structures in the outer flow perturbations influence the roll-up process,
increasing the variance in the size of the shed vortices. This interaction of the large-scale turbulent structures
in the freestream with the roll-up process also causes a greater degree of variability in the shedding period
and wavelength. Furthermore, the higher turbulent dissipation rate weakens the coherence in the roll-up
vortices, causing them to be more diffused and distorted compared to the baseline turbulence case shown in
Fig. 4. The three-dimensional nature of the shear layer roll-up process for higher Tu may have also con-
tributed to a weaker coherence in the shed vortices (McAuliffe and Yaras 2010; Istvan and Yarusevych
2018). This influence of large-scale structures in the freestream flow on the roll-up vortices is better
observed in videos uploaded as supplementary files along with this manuscript.

4 Conclusions

The smoke visualization technique was used to investigate the topology of laminar separation bubbles
(LSBs) over an SD7003 airfoil under the influence of elevated background turbulence levels (Tu ¼ 0:24%),
and comparisons are made with the baseline turbulence case, Tu ¼ 0:02%. Experiments were carried out for
a ¼ 4 and a ¼ 8 and Re ¼ 60; 000 and 100,000. The results demonstrate that increasing the Tu to 0:24%
leads to the elimination of the LSB for a ¼ 4 for both Re ¼ 60; 000 and 100,000. Increasing the angle of
attack (a ¼ 8 ) under elevated turbulence level causes the chordwise extent and height of the LSB to shrink.
The roll-up vortices are smaller, distorted, and more diffused than the baseline turbulence case due to the
increased dissipation rate . The large-scale structures in the freestream perturbation interacted with the
shear layer roll-up process, increasing the variance in the shedding period of the rollers and their size.
Increasing Re from 60,000 to 100,000 for a ¼ 8 and Tu ¼ 0:24% does not have any noticeable impact on
the size of the rollers and the wavelength.
Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1007/s12650-
024-01039-0.

Acknowledgements The authors thank the Kenneth M. Molson Foundation and the Natural Sciences and Engineering
Research Council of Canada (NSERC) for funding this research. The authors also highly appreciate Prof. Swetaprovo
Chaudhuri from the University of Toronto, who provided the Photron Fastcam SA5 camera used in the present study.
Amandine Jollant from The École centrale de Lyon, France, who participated in the video editing process during her summer
internship at UTIAS, is also acknowledged. The authors are grateful to Caulan Rupke and Daniel Asadi from the University of
Toronto for assisting in the fabrication of the smoke-wire apparatus. Lastly, Marcus Luberto from the University of Toronto is
acknowledged for his assistance in manufacturing the turbulence grids used in the present study.
Smoke Visualization of the Effect of Freestream Turbulence...

Author Contributions Suraj Bansal contributed towards (a) the design of the experimental setup, including the new smoke-wire
apparatus, airfoil assembly, and turbulence grids, (b) smoke video acquisition, (c) video editing for the supplementary files,
(d) formal analysis and investigation, and (e) the preparation of the original draft of this manuscript. Philippe Lavoie contributed
towards (a) funding acquisition, (b) research supervision, and (c) review of the manuscript. All authors read and approved the final
manuscript.

Funding This research was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) under
grant number RGPIN-2019-07108. A charitable donation from the Kenneth M. Molson Foundation also funded the research.

Declarations

Conflict of interest The authors have no conflict of interest in this research.

References

Arena AV, Mueller TJ (1980) Laminar separation, transition, and turbulent reattachment near the leading edge of airfoils.
AIAA J 18(7):747–753
Bansal S, Lavoie P (2022) Effect of periodic longitudinal gusts on airfoil performance under separated-flow conditions. AIAA
Paper 2022–4104, AIAA AVIATION 2022 Forum. Illinois, Chicago
Bastedo WG Jr, Mueller T (1986) Spanwise variation of laminar separation bubbles on wings at low Reynolds numbers. J Aircr
23(9):687–694
Burgmann S, Schroder W (2008) Investigation of the vortex induced unsteadiness of a separation bubble via time-resolved and
scanning PIV measurements. Exp Fluids 45:675–691
Burgmann S, Dannemann J, Schroder W (2008) Time-resolved and volumetric PIV measurements of a transitional separation
bubble on an SD7003 airfoil. Exp Fluids 44:609–622
Carmichael BH (1981) Low Reynolds number airfoil survey. Tech. rep., NASA-CR-165803 1
Dellacasagrande M, Barsi D, Lengani D et al (2020) Response of a flat plate laminar separation bubble to Reynolds number,
free-stream turbulence and adverse pressure gradient variation. Exp Fluids 61(128)
Dellacasagrande M, Barsi D, Lengani D et al (2024) Instability processes in short and long laminar separation bubbles. Exp
Fluids 65:109
Drela M (1989) XFOIL: An analysis and design system for low reynolds number airfoils. In: Mueller TJ (ed) Low Reynolds
number aerodynamics. Springer, Berlin, Heidelberg, pp 1–12
Fuchs A, Kharche S, Patil A et al (2022) An open source package to perform basic and advanced statistical analysis of
turbulence data and other complex systems. Phys Fluids 34(10):101801
Hain R, Kahler CJ, Radespiel R (2009) Dynamics of laminar separation bubbles at low Reynolds number aerofoils. J Fluid
Mech 630:129–153
Istvan MS, Yarusevych S (2018) Effects of free-stream turbulence intensity on transition in a laminar separation bubble formed
over an airfoil. Exp Fluids 59(52):1–21
Jaroslawski T, Forte M, Vermeersch O et al (2023) Disturbance growth in a laminar separation bubble subjected to free-stream
turbulence. J Fluid Mech 956:A33
Jones LE, Sandberg RD, Sandham ND (2008) Direct numerical simulations of forced and unforced separation bubbles on an
airfoil at incidence. J Fluid Mech 602:175–207
Kirk TM, Yarusevych S (2017) Vortex shedding within laminar separation bubbles forming over an airfoil. Exp Fluids
58(43):1–17
Kumar R, Sarkar S (2024) Characterizing disturbance growth and transition of laminar separation bubbles subjected to varying
freestream turbulence. Phys Fluids 36(10):104105
Kumar R, Singh P, Sarkar S (2022) Transition of a laminar separated boundary layer under varying adverse pressure gradient.
ASME Turbo Expo, (2022): Turbomachinery Technical Conference and Exposition. Rotterdam, The Netherlands
Kurelek JW, Lambert AR, Yarusevych S (2016) Coherent structures in the transition process of a laminar separation bubble.
AIAA J 54(8):2295–2309
Lambert AR, Yarusevych S (2017) Characterization of vortex dynamics in a laminar separation bubble. AIAA J 55(8):2664–
2675
Lengani D, Simoni D, Ubaldi M et al (2014) POD analysis of the unsteady behavior of a laminar separation bubble. Exp Therm
Fluid Sci 58:70–79
Lissaman PBS (1983) Low Reynolds number airfoils. Annu Rev Fluid Mech 15(1):223–239
Marxen O, Lang M, Rist U (2013) Vortex formation and vortex breakup in a laminar separation bubble. J Fluid Mech 728:58–
90
McAuliffe BR, Yaras MI (2010) Transition mechanisms in separation bubbles under low- and elevated-freestream turbulence. J
Turbomach 132(1):011004 1-10
Mueller TJ, Batill SM (1982) Experimental studies of separation on a two dimensional airfoil at low Reynolds numbers. AIAA
J 20(4):457–463
Mueller TJ, DeLaurier JD (2003) Aerodynamics of small vehicles. Annu Rev Fluid Mech 35(1):89–111
Ol MV, McAuliffe BR, Hanff ES et al (2005) Comparison of laminar separation bubble measurements on a low Reynolds
number airfoil in three facilities. AIAA Paper 2005-5149, 35th AIAA Fluid Dynamics Conference and Exhibit, Toronto,
Ontario
Olson DA, Katz AW, Naguib AM et al (2013) On the challenges in experimental characterization of flow separation over
airfoils at low Reynolds number. Exp Fluids 54(2):1470–1480
 S. Bansal, P. Lavoie

Simoni D, Lengani D, Ubaldi M et al (2017) Inspection of the dynamic properties of laminar separation bubbles: freestream
turbulence intensity effects for different Reynolds numbers. Exp Fluids 58(66):1–14
Wilson PG, Pauley LL (1998) Two- and three-dimensional large eddy simulations of a transitional separation bubble. Phys
Fluids 10(11):2932–2940
Zilli J, Sutton D, Lavoie P (2017) Effect of freestream turbulence on laminar separation bubbles and flow transition on an
SD7003 airfoil at low Reynolds numbers. AIAA Paper 2017-0302, 55th AIAA Aerospace Sciences Meeting, Grapevine,
Texas

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional
affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement
with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely
governed by the terms of such publishing agreement and applicable law.