Content uploaded by Thomas Baheru
Author content
All content in this area was uploaded by Thomas Baheru on Jun 27, 2015
Content may be subject to copyright.
Content uploaded by Thomas Baheru
Author content
All content in this area was uploaded by Thomas Baheru on Jun 27, 2015
Content may be subject to copyright.
Content uploaded by Thomas Baheru
Author content
All content in this area was uploaded by Thomas Baheru on Jun 27, 2015
Content may be subject to copyright.
The 12th Americas Conference on Wind Engineering (12ACWE)
Seattle, Washington, USA, June 16-20, 2013
Wind-Driven Rain Deposition on Building Façade: Development of Test-
Based Data
Thomas Baheru
1
, Arindam Gan Chowdhury
2
, Jean -Paul Pinelli
3
1
PhD Candidate, Civil & Environ. Engineering, Florida Int’l University, Miami, FL, USA, tbaheru@FIU.edu
2
Associate Prof., Civil & Environ. Engineering, Florida Int’l University, Miami, FL, USA, chowdhur@FIU.edu
3
Professor, Civil Engineering, Florida Institute of Technology, Melbourne, FL, USA, pinelli@FIT.edu
Abstract: Wind-driven rain is one of the main sources of damage to building interior and
contents during hurricane landfall. Recently, vulnerability models for hurricane induced total
building interior damage (damage to building interior components, utilities, and contents) have
been widely developed for prediction of property loss in relation to determination of insurance
premiums and development of damage mitigation techniques and guidelines. However, the
prediction capabilities of these models have been limited due to the lack of field and
experimental data on some specific model parameters. This paper presents an experimental study
being conducted to develop test-based data on wind-driven rain intrusion through building
envelope breaches caused by strong wind and wind-borne-debris. The study aimed at
investigating parameters such as rain admittance factor (RAF) and surface runoff coefficient for
different wind directions. The dataset may be used to increase the accuracy and prediction
capabilities of existing catastrophe (CAT) models.
Key words: Building envelope, Hurricane induced damage, Large-scale test, Wind speed, Wind-
driven rain
2
1.
INTRODUCTION
Wind-driven rain is one of the main sources of damage to building interior and contents during
hurricane landfall. Recently, vulnerability models for hurricane induced total building interior
damage (damage to building interior components, utilities, and contents) are being widely
developed for prediction of property loss in relation to determination of insurance premiums and
development of damage mitigation techniques and guidelines (Dao and Lindt, 2010; Pita et al,
2012). However, the prediction capabilities of these models have been limited due to the lack of
field and experimental data on specific model parameters. This paper presents the experimental
study conducted to develop test-based data on wind-driven rain intrusion through building
envelope breaches caused by strong wind and wind-borne-debris. The study aimed at
investigating the mechanisms by which rainwater ingresses to building interior regions through
envelope defects, openings, and breaches and providing test-based data of model parameters for
better estimation of rainwater intrusion.
The vulnerability assessment of building interior components and contents for hurricane induced
damage requires a volumetric estimation of rainwater intrusion through building envelope
defects and breaches. For a given type of envelope opening (breach, opening or defect), the rate
of wind-driven rain intrusion depends on the opening size, location of the opening on the
building envelope, wind-driven rain rate, and wind direction. The rainwater ingress can be either
due to direct impinging raindrops falling onto the opening area or as surface runoff flowing over
the undamaged upstream area of the building envelope. The impinging raindrop distribution is
determined through the rain admittance factor (RAF) which quantifies the amount of rainwater
deposition on the building façade as a function of building shape, location on building façade,
and wind direction(Straube and Burnett, 2000). The final destination of raindrops on a building
façade is also dependent on the raindrop size, implying the importance of raindrop size
distribution in measurements of RAF. Similarly, the surface runoff water at different location on
the building façade is characterized by the surface runoff ratio (Blocken and Carmeliet, 2012).
The surface runoff rainwater over the building façade is affected by the absorption capacity of
building envelope material, friction between water film and building surface, surface tension,
gravity, and the wind action. The wind action is more important during hurricane conditions,
allowing the pressure distribution over the building surface to direct the rainwater in certain
patterns.
The rate of wind-driven rain deposition on a building facade has been studied through field
measurements, wind tunnel experimentation, and using computational fluid dynamics (CFD)
modeling (Blocken and Carmeliet, 2005; Ge and Krpan, 2009; Inculet, 2001; Nore et al, 2007;
Van-Mook, 2002).Blocken and Carmeliet (2005) reported field measurement data of wind-driven
rain on the VIELT building of Laboratory of Building Physics at Katholieke Universiteit Leuven.
The dataset were developed with the intention of providing a high-resolution database for model
development and validation. The study also reported the use of collected wind-driven rain data
for the evaluation of the wind-driven rain coefficients and development of numerical models.
Inculet (2001) measured the impact pattern of raindrops using wind tunnel measurement of
raindrop deposition on a building façade. Results reported from wind-tunnel measurement
demonstrated a raindrop impact pattern similar to numerical models and other field
measurements.
3
Nore et al (2007) measured the wind-driven rain deposition on three low-rise buildings to
provide data for model development and validation purposes. The measurements reported by
Nore et al (2007) contained detailed meteorological measurements of wind and rain both on free-
field and on building facades. The data collected were used to determine the distribution of
raindrop accumulation on the buildings’ façades. The variation of wind direction during the
wind-driven rain event was noted as an inherent challenge in assigning the ratios to a specific
wind angle of attack. The wind direction for a measured rain distribution pattern on the
buildings’ façade was determined by taking weighted average of wind direction based on wind
speed and horizontal rain rate. Van-Mook (2002) also measured the rainwater accumulation on
the west façade of the main building at the Eindhoven University of Technology (TUE). Driving-
rain ratios were estimated at two locations on the building façade. Results showed wide
dispersion of driving-rain ratios (50% of the mean values) as a function of reference wind
velocity and horizontal rain rate (Van-Mook, 1999, 2002).
A more extensive study of building exposure to wind-driven rain was conducted on five low-rise
and three high-rise buildings located in British Columbia by Ge and Krpan (2009). The objective
of the study was to improve the adequacy and use of a wind-driven rain index (wind-driven rain
map) in the estimation of rain exposure during the design of building envelopes. The field
measurement data suggested that wind-driven rain parameters such as catch ratios and/or wall
factors (an alternative term for RAF) are not constant, but could vary with rain events (Ge and
Krpan, 2007). The study also indicated that the presence of a 130mm overhang significantly
reduced the catch ratios and wall factors on the building façade.
Some early-time comprehensive review of wind-driven rain field measurements on building
façades was published by Blocken and Carmeliet (2004) and Straube (1998). These field
measurements and estimated RAF values demonstrated that the top edge and corners of a
building are usually exposed to higher raindrop concentrations, which is largely attributed to the
deflection of raindrops caused by the driving wind force as the consequence of the presence of
the building itself.
Section 2 covers methodology followed to acquire test-based data of wind-driven rain deposition
on building façade with detailed descriptions of testing setup, simulation of tropical cyclone
wind-driven rain, instrumentation, and testing protocol. Section 3 discusses the test results
followed by some important deductions. The conclusions of the study along with summarized
major findings are presented in Section 4.
2.
METHODOLOGY
2.1.
Wind-driven rain simulation
Prior to collecting test data of RAF and surface runoff rainwater on building façades, typical
hurricane wind and wind-driven rain characteristics were simulated in the experimental setup to
ensure a realistic representation in the test-based data for it to be applicable to the development
of hurricane induced loss prediction models. Turbulent wind flow was generated using the newly
built 12-fan Wall of Wind facility at Florida International University (FIU). Vertical spires and
floor roughness elements were used to simulate the target atmospheric boundary layer mean
wind speed and turbulence intensity profiles. Preliminary validation of the wind flow was
conducted through comparison of pressure measurements on building models to those obtained
in the field and in conventional wind tunnel (Aly et al, 2011). In addition to the wind simulation,
4
the raindrop size distribution (RSD) associated with tropical cyclone rain was simulated using
agricultural spray nozzles placed behind the spires. The nozzle types were selected based on the
range of drop sizes that they produce as compared to the target RSD derived from data collected
during 2004 hurricane season. Figure 1 shows the target non-dimensional RSD curve and
statistical values of gamma distribution parameters using combined RSD datasets from
Hurricane Alex, Charley, and Gaston (2004). The field RSD data were acquired through Tropical
Rainfall Measuring Mission (TRMM) at Wallops station, Wallops Island, Virginia, during
Hurricane Alex (2004), Charley (2004), and Gaston (2004).
Figure 1: Raindrop Size Distribution (RSD) of tropical cyclone rainfall.
The normalized RSD shown in Fig. 1 was used to simulate wind-driven rain at different rain
rates. The target wind-driven rain rate (sometimes referred as vertical rain rate: the mass flux of
raindrops passing through a unit vertical area) was determined based the vertically falling rain
rate spectrum during hurricanes as suggested by Lonfat et al (2004). For the selected falling rain
rate, the normalized RSD was converted to dimensional form using the liquid water content
relationship derived used the combined dataset. Assuming no slippage of wind flow over
raindrops surface area, the lateral velocity of the raindrops was considered as the target wind
speed and combined to the dimensional RSD to calculate the target wind-driven rate at that wind
speed. Figure 2 presents the target wind-driven rain rate plot as a function of wind speed
determined based of the RSD of Fig. 1 and a vertically falling rain rate of 25.4 mm/hr (1.0 in/hr).
5
Figure 2: Target WDR rate for vertically falling rain rate (horizontal rain rate, RR
h
) of 25.4 mm/hr (1.0 in/hr).
Once the target wind-driven rain rate and wind speed relationship was developed for the
specified target RSD, the RSD was simulated in the experimental setup by manipulating the
spray rate and the flow pressure using different types of spray nozzles. The simulation result was
compared to the target RSD at various target wind speeds. Fig. 3 demonstrates the RSD
simulation result at a simulated wind speed of 20 m/s (45 mph) and target wind-driven rain rate
of 89 mm/hr (3.5 in/hr). The simulation was conducted at length scale of 1:4 and a velocity scale
of 1:2. Detailed studies of wind-driven rain parameters at various simulation scales and scaling
similarity requirements have been presented by Baheru et al (2012) and Inculet (2001).
Figure 3: Raindrop Size Distribution (RSD) simulation using 12-fan Wall of Wind.
40 60 80 100 120 140 160
0
2
4
6
8
10
12
14
16
18
Wind Speed, U (mph)
Wind-driven rain rate, RR
v
(in/hr)
40 60 80 100 120 140 160
0
2
4
6
8
10
12
14
16
18
Wind Speed, U (mph)
Wind-driven rain rate, RR
v
(in/hr)
6
It is worth to mention that in the RSD simulation (Fig. 3), the number counts of larger size drops
(drop diameter greater than 1.0mm) were rejected because of two reasons: i) It is unlikely for a
nozzle with median-volume diameter of 0.218 – 0.349 mm to produce a drop size larger than 1.0
mm given that the volume contribution increases with drop diameter to the third power, and, ii)
The cumulative total volume of water from all nozzles based on the PIP drop number count was
found to be much larger than the expected value obtained from flow rate and pressure
calculations
2.2.
Test-setup of wind-driven rain intrusion tests
The development of test based data for wind-driven rain intrusion through building envelopes
was conducted on building models with three different types of roof: gable, flat, and hip roof
buildings. The building models were built with a length scale of 1:4 and model-scale dimensions
of 1.5 m wide 2.3 m long and 0.9 m mean roof height. The model scale was selected based on the
Wall of Wind flow field, considering a maximum blockage of 8%. Most of the tests were
conducted with a mean wind speed of 26.8m/s (60 mph) measured at the buildings’ mean roof
height. Three wind directions were considered: 0°, 45°, and 90°, which covered all the possible
45° wind directions taking the advantage of building model symmetry. A total of 38 tests were
conducted on the three building models, collecting either direct impinging rain or surface runoff.
Table 1 summarizes the test protocol. The direct impinging rainwater on gable roof building was
measured for RAF at three different wind speeds keeping the same wind direction to test the
hypothesis that the RAF value at a given location on a building façade is independent of the wind
speed. The measurements were conducted at a wind-driven rain rate of 330.2 mm/hr (13.0 in/hr)
except for the RAF study at different wind speeds. At lower wind speeds the wind-driven rain
rate was lower as discussed in the previous section (see Fig. 2)
Table 1: Testing protocol for Rain Admittance Factor (RAF) and Surface Runoff (SR) tests
Building Type
Test
Type Bucket
Loc.
Wind Speed
30mph,
45mph
Wind Speed
60mph
Rain Rate
13.0in/hr
Wind Direction Number of
tests
Gable roof bldg
RAF √ √ √
**
0°(360°) 3
Gable roof bldg
RAF √ √ 45°, 90° 2
Hip roof bldg RAF √ √ 0°(360°), 45°, 90°
3
Flat roof bldg RAF √ √ 0°(360°), 45°, 90°
3
Gable roof bldg
RAF+SR
1,2,3,4,5,6
√ √ 0°(360°), 45°, 90°
18
Flat roof bldg RAF+SR
1,2,3 √ √ 0°(360°), 45°, 90°
9
**
The wind-driven rain rate for wind speeds of 30 and 45mph were different from 13.0in/hr (see Fig. 2)
Plastic rain collecting buckets developed for the studies were mounted on the roof and walls of
the building models to collect the direct impinging raindrops. Each bucket developed for the
RAF tests was provided with a rim around the periphery of the bucket’s catch area to collect rain
water due to impinging raindrops only and avoid collecting the surface runoff. To collect surface
runoff rain water together with direct impinging raindrops, the edges of each runoff bucket were
made flush (i.e., edges without rim) with the building façade. The buckets for the later tests were
mounted in a row at specific height on the building façade. The tests were conducted for six rows
along the height of the building measuring surface runoff water at specific height at a time. Rain
water volume due to the surface runoff was determined as the difference of rain water volume
7
measured using the two types of buckets. Graduated beakers placed inside the building were
connected to each bucket to collect the simulated rainwater. Measurement by weight was used to
record the amount of water in the beaker.
3.
TEST RESULTS AND DISCUSSION
The measured volume of direct impinging raindrops and surface runoff rainwater were
normalized with the wind-driven rain rate and bucket catch area to determine the RAF and
normalized surface runoff coefficients (SRC) at different locations on the building façade for
wind direction. For the roof of the buildings, the RAFs and SRCs were determined using a
horizontal rain (vertically falling rain) rate of 25.4 mm/hr (1.0 in/hr).
0.2
0.2
0.4
0.4
0.6
0.6
0.8
0.8
1
1
0
3.3329e-008
0.2
0.2
0.2
0.2
0.2
0.4
0.4
0.4
0.4
0.6
1
0
°
8
Figure 4: Rain admittance factors (RAF) for gable roof building model with 0°, 45°, and 90° wind angle of attack.
Figure 4 shows the distribution of direct impinging raindrops (RAF) on the wall and roof of the
gable-roof building model for various wind angles of attack. The test result indicated that the
windward faces of the building received high concentration of rainwater while the leeward faces
were protected from direct impinging raindrops. The formation of flow separation and
0.2
0.3
0.3
0.4
0.4
0.5
0.6
0.7
0.1
0.2
0.3
0.3
0.4
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
0.1
0.1
0.1
0.1
0.2
0.2
0.2
0.2
0.3
0.3
0.3
0.3
0.4
0.5
1
3.7474e-008
0.1
0.1
0.1
0.2
0
20
40
60
80
100
120
0.2
0.2
0.4
0.4
0.6
0.8
0.8
1
1
1.2
1.7972e-007
0.2
0.2
0.2
0.2
0.2
0.4
0.4
0.4
0.6
1
1.8297e-007
0.2
0.2
0.4
0.4
0.6
0.8
1
1.2
1.7972e-007
0.2
0.2
0.2
0.2
0.2
0.4
0.4
0.4
0.6
1
4.6494e-008
0.2
90
°
45
°
9
development of negative pressure at the leeward façades of the building minimized the
deposition of raindrops. Generally the lower values of rain admittance factor (RAF) were
observed at the bottom of the building and its value increased with the height. Despite the
presence of roof overhang, higher magnitudes of RAF were noticed at the top part of the building
walls indicating the effect of high wind speeds. However, the overhang protected the building
walls from rainwater flowing over the roof of the building as surface runoff.
Results of the RAF investigation at various wind speeds indicated minor differences in
magnitude, which could be attributed to the change in the projectile of the raindrops as the wind
speed changes. At higher wind speeds, raindrop particles tend to cover longer horizontal
distances for the same vertical falling distance when compared to low wind speeds. Repeated
measurements on three building models indicated that rainwater deposition on the windward
faces was dominantly due to the direct impinging raindrops while the accumulation of rainwater
on the leeward faces was mainly due to surface runoff.
4.
CONCLUSIONS
A series of wind-driven rain tests were conducted on low-rise building models to develop test-
based data of rain admittance factors (RAF) and SRCs. The test-based rainwater intrusion data
were developed with the intention of providing experimental data for interior damage prediction
model development. The dataset will be used to increase the accuracy and prediction capabilities
of existing CAT models.
ACKNOWLEDGMENTS
The research team would like to thank the sponsor of the project, the Florida Sea Grant College
Program (FSGCP). We thank Dr. Tokay Ali, NASA Goddard Space Flight Center, Greenbelt,
Maryland for providing RSD data collected during hurricane. We thank Dr. Forrest Masters of
University of Florida for providing a precipitation imaging probe used to measure simulated
wind-driven rain. We acknowledge Dr. Bert Blocken of Building Physics and Services,
Eindhoven University of Technology, for his helpful comments and advice on the project.
The opinions, findings, and conclusions presented in this article are those of the authors alone,
and do not necessarily represent the views of the sponsoring agency.
The 12th Americas Conference on Wind Engineering (12ACWE)
Seattle, Washington, USA, June 16-20, 2013
REFRENCES
Aly, A.M., Chowdhury, A.G., Bitsuamlak, G., 2011. Wind profile management and blockage assessment
for a new 12-fan Wall of Wind facility at FIU. Wind and Structures 14, 1-16.
Baheru, T., Chowdhury, A.G., Bitsuamlak, G., Tokay, A., 2012. A parametric representation of wind-
driven rain in experimental setups, in: Proceedings of Applied Technology Council and Structural
Engineering Institute Conference (ATC & SEI 2012) Advances in Hurricane Eingineering: Learning
from our past, 270-282. ASCE and ATC 2013.
Blocken, B., Carmeliet, J., 2004. A review of wind-driven rain research in building science. J. Wind Eng.
Ind. Aerodyn. 92, 1079–1130.
Blocken, B., Carmeliet, J., 2005. High-resolution wind-driven rain measurements on a low-rise
building—experimental data for model development and model validation. J. Wind Eng. Ind.
Aerodyn. 93, 905-928.
Blocken, B., Carmeliet, J., 2012. A simplified numerical model for rainwater runoff on building facades:
possibilities and limitations. Building and Environment CDROM.
Dao, T.N., Lindt, J.W.v.d., 2010. Methodology for wind-driven rainwater intrusion fragilities for light-
frame wood roof systems. Journal of Structural Engineering, ASCE 136, 700–706.
Ge, H., Krpan, R., 2007. Field measurement of wind-driven rain on a low-rise building in the coastal
climate of British Columbia, in: Proceedings of the 11th Canadian Conference on Building Science
and Technology.
Ge, H., Krpan, R., 2009. Wind-driven rain study in the coastal climate of British Columbia. British
Columbia Institute of Technology, Burnaby, British Columbia, pp. 1-51.
Inculet, D.R., 2001. The design of cladding against wind-driven rain, in: Proceedings of the PhD
dissertation.
Lonfat, M., Frank D. Marks, J., Chen, S.S., 2004. Precipitation Distribution in Tropical Cyclones Using
the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager: A Global Perspective. Monthly
Weather Review, American Meteorological Society 132, 1645-1660.
Nore, K., Blocken, B., Jelle, B.P., Thue, J.V., Carmeliet, J., 2007. A dataset of wind-driven rain
measurements on a low-rise test building in Norway. Building and Environment 42, 2150–2165.
Pita, G.L., Pinelli, J.-P., Cocke, S., Gurley, K., Mitrani-Reiser, J., Weekes, J., Hamid, S., 2012.
Assessment of hurricane-induced internal damage to low-rise buildings in the Florida Public Hurricane
Loss Model. Wind Eng. Ind. Aerodyn. 104-106, 76-87.
Straube, J.F., 1998. Moisure Control and Enclosure Wall Systems, in: Proceedings of the PhD
dissertation.
Straube, J.F., Burnett, E.F.P., 2000. Simplified prediction of driving rain on buildings, International
Building Physics Conference, Eindhoven, Netherlands, pp. 375-382.
Van-Mook, F.J.R., 1999. Full-scale measurements and numeric simulations of driving rain on a building,
the 10th International Conference on Wind Engineering, Kobenhavn, pp. 21-24.
Van-Mook, F.J.R., 2002. Driving rain on building envelopes. Ph.D. Thesis, Building Physics Group
(FAGO) Eindhoven University of Technology, Eindhoven University Press, Eindhoven, the
Netherlands.
