Research Article

Transportation Research Record

1–13
Ó National Academy of Sciences:
Optimizing the Performance of Transportation Research Board 2019
Article reuse guidelines:
Microsurfacing Treatments in Flexible sagepub.com/journals-permissions
DOI: 10.1177/0361198118821874

Pavements and Assessing Its Effects on journals.sagepub.com/home/trr

Moisture Damage

Mohammad Z. Bashar1, Mostafa A. Elseifi1, Momen R. Mousa1,

Zhongjie Zhang2, and Kevin Gaspard2

Abstract
Louisiana’s $6.3 million microsurfacing program is amongst the largest microsurfacing programs in the United States. As
microsurfacing seals the road surface, the effectiveness of this treatment in such a setting has been a concern in recent years
by linking it to moisture damage caused by the trapped moisture underneath the pavement especially in areas with shallow
groundwater table. Furthermore, the cost-effectiveness and optimal timing of microsurfacing applications are also not well
established for the south-central United States. The objective of this study was to assess the short-term and long-term
effects of microsurfacing treatments and to evaluate whether microsurfacing is a major contributor to moisture damage.
Field performance of 27 sections where microsurfacing treatments were applied between 2003 and 2008 was monitored for
at least eight years. Results indicated that microsurfacing is most effective in addressing rutting damage as compared with the
other performance indicators. Microsurfacing extended the service life of the pavements by 4.9–8.8 years. The effectiveness
of microsurfacing was found to be optimum when applied to pavements with pre-treatment conditions ranging from 80 to
85. Results of this study also showed that microsurfacing-treated sections exhibited higher percentage of moisture damage as
compared with the untreated sections in all the districts. Therefore, an in-depth assessment of the effects of microsurfacing
on moisture damage in asphalt pavements is recommended.

Microsurfacing is a mixture of crushed aggregate, asphalt year make the pavements highly vulnerable to water
emulsion, water, polymer additive, and mineral fillers entrapment and moisture damage. As microsurfacing
used to correct wheel ruts, improve surface friction, and seals the road surface completely, the effectiveness of this
extend pavement service life (1, 2). It is usually a single treatment in such a setting has been a concern in recent
aggregate thick and is placed as a thin lift of mixture typi- years by linking it to moisture damage caused by the
cally containing 82–90% aggregate, 1.5–3.0% mineral fil- trapped water underneath the surface especially in areas
ler, and 5.5–9.5% residual asphalt (3, 4). Microsurfacing with shallow GWT. Furthermore, the cost-effectiveness
has been found to be equally effective for both low- and and optimal timing of microsurfacing applications are
high-volume roadways. It also allows rapid opening of also not well established for the south-central United
the roads to traffic, which makes it an effective preven- States. Proper knowledge of performance and cost com-
tive maintenance technique (5, 6). The major challenge in ponents of the surface treatments under these conditions
microsurfacing is that it requires special equipment for will allow the associated pavement maintenance agencies
application, which makes it an expensive form of preven-
tive maintenance as compared with other common sur-
face treatments such as chip seal or slurry seal treatment 1
Department of Civil and Environmental Engineering, Louisiana State
(3). University, Baton Rouge, LA
2
The benefits of microsurfacing have been well Louisiana Transportation Research Center, Louisiana State University,
Baton Rouge, LA
researched and documented in previous studies (2, 3, 7).
However, in Louisiana, shallow groundwater table Corresponding Author:
(GWT) and heavy rainfall conditions throughout the Address correspondence to Mostafa A. Elseifi: elseifi@lsu.edu
2 Transportation Research Record 00(0)

to set up a more reliable schedule and realistic budget for conventional dense-graded resurfacing (15). The study
the maintenance and rehabilitation activities (7). estimated that microsurfacing resulted in 5–7 years
increase in service life for the southeastern region of the
U.S., where the cost of microsurfacing mix ranged from
Objectives $1.07 to $1.20/m2. Hixon and Ooten showed that the ini-
This objective of this study was twofold. First, the short- tial cost of microsurfacing is 55% of an overlay; yet the
and long-term performances of microsurfacing were eval- annual cost of microsurfacing was found to be slightly
uated as related to the pre-treatment conditions of the higher than an asphalt concrete overlay (11). Most of the
pavement. Performance curves were developed to assess past research pertaining to microsurfacing has mainly
the cost-effectiveness of the treatment. Effectiveness focused on the general effectiveness of the treatment and
models in conjunction with the cost-benefits of the proj- did not take into account exogenous factors such as spe-
ects were used to identify the optimum timing for this cial geographical and climatological issues. In addition,
preventive maintenance activity. Second, treated sections past studies did not evaluate the effects of microsurfacing
were evaluated to assess whether microsurfacing signifi- on moisture damage.
cantly contributes to moisture damage.

Methodology
Background
Flexible pavements with microsurfacing treatment and
Louisiana’s $6.3 million microsurfacing program is that satisfied other project acceptance criteria, detailed
amongst the largest microsurfacing programs in the subsequently, were selected for a detailed performance
United States (2). Louisiana has a humid subtropical cli- and economic evaluation. Performance curves for each
mate, characterized by long, hot, humid summers with of the sections were developed by modeling pavement
heavy rainfalls throughout the year (average annual rain- conditions as a function of time. Long-term pavement
fall of 60 in.) in which microsurfacing treatments primar- performance data were used to quantify the benefits of
ily serve the purpose of waterproofing the pavement microsurfacing treatments in terms of pavement perfor-
surface. Temple et al. investigated the performance of mance indicators.
Louisiana’s microsurfacing program in 2002 (8). The
study analyzed treated sections for 60 months and
detected significantly fewer cracks and substantial reduc- Pavement Performance Indicators
tion in rutting after treatment. However, this study
Louisiana pavement management system monitors
assessed the effectiveness of microsurfacing holistically
roughness, rutting, and cracking (alligator, random,
and did not provide insights into the factors affecting the
transverse, and longitudinal), which are the critical per-
performance of microsurfacing in Louisiana.
formance indicators in terms of safety, ride quality, and
Labi et al. studied the long-term performance of
microsurfacing applications in Indiana where it was overall condition of the pavement. However, not all the
found to extend the service life of the pavement by 4–15 indicators are equally important for different treatment
years (7). Most of the studies have reported a treatment types. Therefore, in selecting the most appropriate indices
service life of 4–7 years if microsurfacing is applied prop- for a treatment effectiveness evaluation, it is vital to con-
erly (8–12). Microsurfacing was found to be most sider the type and the existing conditions of the pavement
effective in addressing rutting as it reduced rutting by along with the method of evaluating the performance of
90–96%, whereas it was only 7–27% effective in reducing the treatment (7).
the surface roughness of the pavement (7, 13). A study Microsurfacing is typically applied to treat rutting,
conducted by the National Center for Asphalt roughness, and surface irregularities. Indices that reflect
Technology (NCAT) evaluated the field performance of the changes in extent and severity of these distresses
23 preservation treatments including fog seals, crack would be the ideal performance indicators for microsur-
seals, chip seals, cape seals, microsurfacing, plant mix facing. Therefore, to investigate the effectiveness of
overlays, and ultrathin bonded wearing course on a local microsurfacing treatments, the following performance
asphalt road in Alabama with a high percentage of indicators were considered in this study:
trucks (14). The use of microsurfacing in combination
with crack sealing was found to very effective against the  Rutting index (RTI): a measure of the average
development of interconnected cracks. longitudinal depressions in the pavement wheel
Watson and Jared evaluated the experience of paths;
Georgia Department of Transportation (DOT) with  Roughness index (RFI): a measure of the longitu-
microsurfacing as an economical alternative to dinal irregularities in the pavement surface;
Bashar et al 3

 Pavement condition index (PCI): represents the Measures of Effectiveness

overall conditions of the pavement. The initial effects, if any, of microsurfacing on rough-
ness, rutting, and overall conditions of the pavement
All the indices are measured on a scale that ranges from were evaluated in terms of performance jump and dete-
0 to 100, where 100 represents the best possible rioration rate reduction. Long-term effects of the treat-
condition. ment were evaluated in terms of effectiveness and service
life extension.

Pavement Performance as a Function of Time

Performance Jump (PJ). PJ is the immediate improvement
Several studies have used a polynomial approach to in pavement conditions after applying the surface treat-
model pavement conditions (3, 16, 17). In this study, it ment (18). Pavement conditions just before and after the
was assumed that both pre- and post-treatment condi- treatment can be estimated as:
tions could be represented by polynomial models (see
Figure 1) as: fpre ðti Þ = a1 ti2 + b1 ti + c1 ð3Þ
fpre ðtÞ = a1 t2 + b1 t + c1 ð1Þ fpost ð0Þ = c2 ð4Þ

fpost ðtÞ = a2 t2 + b2 t + c2 ð2Þ PJ is calculated as the difference in condition index

values just before and after the treatment and therefore,
where was estimated by subtracting Equation 3 from Equation
a1 , b1 , c1 , a2 , b2 and c2 = fitting parameters related to 4 as:
pavement conditions and deterioration rates over  
time for pre- and post-treatment performance models; PJ = c2  a1 ti2 + b1 ti + c1 ð5Þ
and t = time in years.
According to the models, the conditions of a pave-
ment will deteriorate over time following curve AC as Deterioration Rate Reduction (DRR). DRR refers to the slow-
shown in Figure 1; however, if any treatment is applied ing down of the pavement deterioration. The pre- and
at time ti (point B), the PCI will increase to point D. This post-treatment rate of deterioration of the pavement is
immediate increase in PCI following a treatment activity given by the slopes of the performance curves just before
is known as a performance jump. After the jump, the and after the treatment (19) as:
deterioration pattern will follow the curve DE. The time 
is set equal to zero at point D for the post-treatment per- dfpre ðtÞ
Rpre = = 2a1 ti + b1 ð6Þ
formance curve. dt t = ti

Pre-treatment performance curve

A D Rpost Post-treatment performance curve

PJ
Performance Indicator

B
Rpre
A2
A1
C E
Threshold
ti

SLpre SLE

SLpost

Pavement age

Figure 1. Pre- and post-treatment performance curves due to microsurfacing application.

4 Transportation Research Record 00(0)


dfpost ðtÞ SLpost = pavement age with treatment to the
Rpost = = b2 ð7Þ
dt t = 0 threshold;
TV = threshold value for PCI; and
where ti = time of the treatment activity in years.
Rpre and Rpost = rate of deterioration just before and SLE is the increase in pavement service life due to a
after the treatment. treatment application and is given by the difference
DRR is estimated as the difference in pre- and post- between pre- and post-treatment service lives as:
treatment rate of deterioration, that is:
SLE = SLpost  SLpre ð13Þ
DRR = Rpost  Rpre ð8Þ
where
where
DRR = deterioration rate reduction due to a treat- SLE = service life extension in years:
ment application

Effectiveness (E). E is defined as the increase in average Data Collection

pavement conditions over the long term due to microsur-
facing (7). For a treated section, the average pavement Louisiana DOTD (LaDOTD) databases including
conditions over the service life, PAVG , can be obtained as: Pavement Management System (PMS), Highway Needs,
Tracking of Projects (TOPS), Material Testing System
1 (MATT), Letting of Projects (LETS), and Project and
PAVG = ðy 1 + y 2 + . . . + y n Þ ð9Þ
nc Highway Information Inventory were analyzed to collect
the data required for a detailed performance evaluation
where and economic analysis. The pavement condition data,
y1 = pavement condition after treatment;
such as cracking, roughness, patching, and rutting mea-
y2 , y3 , ::yn1 = pavement condition at different years
surements are collected biennially using the Automated
after treatment;
Road Analyzer (ARANÒ) and the results are reported
yc = pavement condition at the end of service life after
every 0.1 miles.
treatment; and
For flexible and composite pavements, the random
nc = number of years the pavement condition was
cracking index encompasses all random cracks, which
measured after treatment.
include thermal, reflective, longitudinal, block, and
Effectiveness is the percentage change in average pave-
cement-treated reflective cracks. A preliminary list of
ment conditions due to microsurfacing relative to the
microsurfacing projects was prepared using the TOPS
pre-treatment conditions of the pavement:
and LETS databases. To determine the exact location,
  length, and the date of the microsurfacing application,
PAVG  PINI
E= 3 100 ð10Þ video logs were reviewed using the VisiWeb software, as
PINI
shown in Figure 2. This software offers geocoded videos
where of the pavements along with other relevant characteris-
PINI = pre-treatment condition of the pavement in tics of the pavements, such as cracking pattern, rough-
terms of performance indicators. ness, and elevation.
In Louisiana, microsurfacing mixture consists of a
uniform mixture of mineral aggregate, polymer modified
Service Life Extension (SLE). The pre- and post-treatment asphalt ranging from 6.0 to 9.0% by dry weight of aggre-
performance curves will reach the threshold at different gate, mineral filler not exceeding 3.0% by dry weight of
times, which can be estimated using: aggregate, water, and additives. The mineral aggregate
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi shall conform to the graduation requirements shown in
b1  b21  4a1 ðc1  TV Þ Table 1. The material is mixed by an auto-sequenced,
SLpre = ð11Þ
2a1 self-propelled micro-surfacing mixing machine capable
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi of accurately delivering and proportioning the materials
b2  b22  4a2 ðc2  TV Þ and discharging the mixed product on a continuous flow
SLpost = + ti ð12Þ
2a2 basis. Ruts and irregularities that are less than 0.5 in. in
where depth are typically covered with a full-width scratch coat
SLpre = pavement age with no treatment to the and ruts exceeding 0.5 in. in depth are filled indepen-
threshold; dently using a rut-filling spreader box having a width of
Bashar et al 5

Figure 2. Pavement images showing the before and after of microsurfacing application on a selected road segment.

Table 1. Aggregate Gradation and Job Mix Formula (JMF) Limits Table 2. Size and Description of the Data Sets Used in the
Analysis
U.S. sieve % Passing JMF tolerances
Number Number of
3/8 inch 100 65 Dataset ID Analysis Index of sections log miles
No. 4 70–90 65
No. 8 45–70 65 1 Field PCI 25 322
No. 16 28–50 65 performance RTI 27 324
No. 30 13–35 65 RFI 27 360
No. 50 10–25 64 2 Moisture PCI 21 25a
No. 100 7–18 63 damage
No. 200 5–15 62
% binder 6 0.5 a
Number of cores.

5–6 feet. For a single-coat application, a spread rate of suitable for the performance analysis. However, fewer
30 lbs./yd2 of dry aggregate is typically used. sections were found to conform to the second criteria, as
To facilitate quantification of treatment effectiveness it is very uncommon for a section not to receive any
and plotting of the pavement performance model, a can- treatment or maintenance activity other than microsurfa-
didate project had to meet the following project accep- cing for a long period of time.
tance criteria: To achieve a higher degree of accuracy in evaluating
the effectiveness of microsurfacing application, each 0.1
 A road segment should have pavement condition mile of the section was used as a data point instead of
data available for a minimum of four cycles before using an average value for the whole section. Table 2
and four cycles after the application of microsur- summarizes the total number of sections and log miles
facing so that non-linear models can be fitted identified for the analysis. Performance curves were
through the data points. developed for each of the data points and threshold val-
 Distress data should follow a decreasing pattern ues depending upon the road classification were used in
over the years except for the year after treatment, the analysis. All the selected sections had a terminal
which should exhibit an increase in the distress threshold value ranging from 50 to 64. For consistency
index value. of the analysis, a threshold value of 60 was assumed in
all the calculations.
The earliest distress data available in the LaDOTD The temperature and precipitation data were
database were collected in 1995, and since then the dis- extracted from the National Oceanic and Atmospheric
tress data have been collected every two years (20). Administration (NOAA) database. The 30-year averages
Hence, ten data sets were available, and the latest avail- were estimated by averaging the climatological values
able data set was recorded in 2015. As the candidate over the period ranging from 1980 to 2010. Geographic
projects should have at least four data points before and coordinates were used to assign temperature and precipi-
four data points after the treatment, sections receiving tation values to each log mile of the treated sections. The
treatments between 2003 and 2008 were found to be mean annual precipitation ranged from 49–68 in. per
6 Transportation Research Record 00(0)

(a)
60
y = -0.52x + 54.02 y = -0.78x + 77.92 y = -0.67x + 65.46
R² = 0.32 R² = 0.81 R² = 0.64
40
PJ

20

0
40 60 80 100 40 60 80 100 40 60 80 100
PCI- RTI- RFI-

(b)
20
y = -0.15x + 15.88 y = -0.25x + 26.76 y = -0.17x + 17.51
15 R² = 0.29 R² = 0.38 R² = 0.59

10
DRR

0
40 60 80 100 40 60 80 100 40 60 80 100
-5
PCI- RTI- RFI-

Figure 3. (a) Performance jump (PJ) and (b) deterioration rate reductions (DRR) as a function of pre-treatment condition of the
pavement.

year and the mean annual temperature varied from greater as compared with the other indices (see Figure 3),
62–71° Fahrenheit. which is in line with the findings from previous studies (7,
The cumulative effects of traffic and climate on pave- 11). For all the log miles, PJ for RTI- showed positive
ment conditions over time were estimated using the traf- values with a mean of 18.1 6 13.9, whereas the mean for
fic (ADTX) and precipitation (AAPX) load factors: PCI- and RFI- ranged from 9.9 to 11.6. Microsurfacing
slowed down the deterioration of rutting condition by 7.2
ADTX = ADT 3 tx ð14Þ units/year, whereas it slowed down the deterioration of
AAPX = AAP 3 tx ð15Þ roughness condition by 3.7 units/year. The correlation
coefficients obtained for PCI- for both cases were found
where to be minimum. This is because the computation of the
ADT = average daily traffic; composite PCI index involves all the condition measures,
AAP = average annual precipitation; and some of which may not get affected due to the applica-
tx = time from a previous major maintenance activity. tion of microsurfacing.

Benefits Quantification Long-Term Effectiveness

Initial Effects Microsurfacing was found to be 10.0 and 11.3% effective
PJ and DRR due to the microsurfacing application were for the rutting and overall condition of the pavement in
calculated using Data Set 1 (see Table 2). Both PJ and the long term according to Equation 10. However, as can
DRR results indicated that microsurfacing was most be seen from Figure 4a, 21.1% of the log miles for PCI-
effective in immediately improving the rutting condition exhibited a negative E value. E as a function of pre-
of the pavement where the slopes were found to be treatment condition of the pavement can be useful in
Bashar et al 7

(a)
100
y = -0.78x + 74.08
75 y = -0.97x + 82.45 R² = 0.54 y = -0.63x + 56.84
R² = 0.62 R² = 0.60
50
E

25

0
40 60 80 100 40 60 80 100 40 60 80 100
-25
PCI- RTI- RFI-

(b)
20.0

15.0
EPCI(%)

10.0

5.0

0.0
0 40 80 120 0 1000 2000 0 3 6 9 12
ADTX (in thousands) AAPX (inch) AC thickness (inch)

Figure 4. Effectiveness of microsurfacing treatment in the long-term.

determining the upper limit beyond which applying a values ranging from 80 to 85. As shown in Figure 5,
treatment will have a negligible effect on improving pave- treatments applied to pavements with PCI-.90 resulted
ment performance. The threshold values were estimated in negative SLE. However, microsurfacing treatments
to be 85, 95, and 90 for PCI, RTI, and RFI, respectively. applied in different geographical locations in Louisiana
The effect of traffic load on the effectiveness shows (Figure 5b) did not show large variation in performance
that E increases up to a certain point and then starts to in terms of SLE, where the mean values ranged from 5.7
drop sharply as the traffic load increases, see Figure 4b. to 7.9 years. The effects of traffic load on the perfor-
Treatment effectiveness is optimized when applied to mance of the microsurfacing treatments were found to
pavements with ADTX less than 70,000 as can be be similar as the effect of ADTX on the long-term effec-
inferred from Figure 4b. Sections receiving precipitation tiveness of the pavement. An optimum SLE of 8.8 years
(AAPX) less than 500 had a significantly higher effec- was obtained when ADTX values ranged from 50,000 to
tiveness as compared with the sections receiving higher 70,000. No definite trend in SLE was observed for vary-
precipitation load, although no definite relationship was ing AC layer thickness and precipitation load.
observed. The effectiveness also increased with increasing
asphalt concrete (AC) layer thicknesses up to a thickness
of 7 in., where an E of 15.8% was obtained. Treatment Cost-Effectiveness (CE)
The CE of a treatment is defined as the ratio of treatment
net benefits (TNB) to the unit cost of the treatment (4). It
Service Life Extension (SLE)
is estimated as:
Microsurfacing was found to be extending the service life
of the pavements by 4.9–8.8 years. SLE in relation to the TNB
CE = ð14Þ
pre-treatment condition of the pavements showed that Unit cost of the treatmentð$=mileÞ
SLE is optimized when applied to a pavement with PCI
8 Transportation Research Record 00(0)

(a) (b)
15 15

10
SLE (years)

SLE (years)
10
5

0 5

-5
70-75

75-80

80-85

85-90
<70

>90
0
2 5 7 8 58
PCI- District

(c)
10
SLE (Years)

0
0 25 50 75 100 0 1000 2000 3000 0 3 6 9 12
ADTX (in thousands) AAPX (inch) AC Thickness (inch)

Figure 5. Service life extension due to microsurfacing application.

TNB is calculated as the increased area under the perfor- A1 = area enclosed between pre-treatment perfor-
mance curve due to the treatment activity. According to mance curve and threshold value.
Figure 1, TNB can be expressed as: The average TNB for microsurfacing treatments was
estimated to be 162:2 6 88:8 PCI-years. The variation of
TNB = A2  A1 ð15Þ TNB with pre-treatment condition is presented in
ð ti
SLpost Figure 6. Cost data for Louisiana indicated that these
  treatments had an average cost of $47,710 per lane-mile.
A2 = fpost  TV dt
As unit cost remained more or less the same for all the
0
a2  3 b2  2   projects, CE followed the trend of TNB when the log
)A2 = SLpost  ti + SLpost  ti + ðc2  TV Þ SLpost  ti miles were compared together based on pavement condi-
3 2
ð16Þ tions at the time of treatment. The performance in terms
of both CE and TNB were found to be optimum for the
SLðpre
  log-miles having pre-treatment PCI values ranging from
A1 = fpre  TV dt 80 to 85, see Figure 6. As observed for the other long-
ti term measures such as E and SLE, the CE values
a1  3 b1  2   dropped below zero for the locations with PCI- . 95.
)A1 = SLpre  ti3 + SLpre  ti2 + ðC1  TV Þ SLpre  ti
3 2
ð17Þ
Treatment Timing Optimization
where
A2 = area enclosed between post-treatment perfor- In Louisiana, surface treatments are usually considered
mance curve and threshold value; when the condition index of a pavement reaches a
Bashar et al 9

in influencing the performance of microsurfacing treat-

1.20 250
ments. Therefore, weighting the results from E, SLE, and
1.00 200 CE, a ranking of these factors was determined to facili-
0.80 150 tate a more accurate treatment timing and selection of

TNB (PCI-Year)
0.60 100 the candidate projects.
CE (x100)

0.40 50
There remains a possibility of error in using PCI as an
optimization criterion as it is a combination of all the
0.20 0
condition indices, some of which may not be affected or
0.00 -50
be negatively affected by a specific type of treatment.
-0.20 -100 Future work could further improve the selection of can-
-0.40 -150 didate projects through investigating the individual pre-
<70 70-75 75-80 80-85 85-90 90-95 95-100
PCI-
treatment distresses such as rutting, roughness, and
cracking indices.
Figure 6. Cost-effectiveness of microsurfacing treatments.
Microsurfacing and Moisture Damage
Table 3. Optimum Conditions for Microsurfacing Treatments in Due to the geographical settings of Louisiana, the pave-
Louisiana ments are highly susceptible to moisture damage. As
microsurfacing seals the pavement surface, it will be logi-
Factors cal to suspect that it may trap moisture underneath the
Pre-treatment AC thickness pavement causing the moisture to cause progressive dam-
Rank condition (PCI) ADT (inch) age in the long term. To assess whether microsurfacing
does contribute to moisture damage, a comparative study
1 80–85 1500–3000 5–7 of the sections with moisture damage was conducted.
2 75–80 \1500 3–5
3 70–75 3000–5000 7–9 LaDOTD performed pavement coring throughout the
state where a total number of 1213 cores were collected
from the different districts in Louisiana. To evaluate if
moisture damage correlates with the geographical loca-
threshold. However, depending upon a single threshold tion of the pavement, the number of stripped cores, iden-
value based on highway classification is highly arguable tified through visual inspection, were quantified for both
as factors such as pre-treatment condition, traffic, AC sections with microsurfacing and untreated sections in
thickness, climate and other exogenous factors influence each district. Table 4 summarizes the findings from the
the performance of these treatment activities signifi- core reports.
cantly. Treatment timing is critical for treatment perfor- The percentage of moisture damage in sections with
mance optimization. microsurfacing was compared with the percentage of
Table 3 summarizes the treatment timing ranking for stripping in untreated sections. The results indicated that
microsurfacing treatments in terms of pre-treatment con- the sections with microsurfacing treatments exhibited
dition of the pavement. ADT and AC thickness of the higher percentage of stripping in all the districts as com-
pavements were also found to be statistically significant pared with the untreated sections; see Figure 7. The

Table 4. Summary of the Cores Extracted from Different Districts in Louisiana

Untreated sections Sections with microsurfacing

District Total cores Number of cores Cores with stripping Number of cores Cores with stripping

2 95 91 4 4 2
3 240 238 54 2 0
4 154 150 24 4 0
5 74 71 25 3 0
7 108 98 49 10 6
8 136 130 40 6 3
58 76 65 15 11 3
61 178 171 48 7 3
62 152 142 13 10 1
10 Transportation Research Record 00(0)

Untreated Microsurfacing
70
% of Stripped sections

60
50
40
30
20
10
0
2 3 4 5 7 8 58 61 62
District

Figure 7. Extent of moisture damage in different districts of Louisiana.

sections with microsurfacing in Districts 2, 7, 8 and 61 Section 193-01

had 10–45% more stripped cores than the untreated sec- The road section is located in Cameron Parish of District
tions. Another key observation from Figure 7 is that
7. It was constructed in 1930 and carries an ADT of
moisture damage in pavements is significantly influenced
1620. The section had a length of 13.38 miles and the
by the geographic location of the pavements regardless
pavement structure consisted of an 8.5 in. AC and 9.5 in.
of its treatment condition. For example, District 7, which
of crushed gravel with a sand base on top of a clay sub-
is generally characterized by its shallow GWT, had the
grade. A microsurfacing maintenance was conducted in
highest percentage of stripped cores for both treated and
2010. At the location of the core, the PCI was 91.4 and
untreated sections. This observation suggests that shal-
the average PCI along the section was 86.4. The section
low GWT could contribute to moisture damage in AC
receives about 62 in. of annual rainfall. The core location
pavements in general (treated or untreated) due to moist-
is within half a mile radius of the Bayou Serpent River,
ure entrapment under the AC layer. To assess further the
which may indicate a shallow GWT.
effects of microsurfacing on moisture damage, 21 sec-
tions with microsurfacing treatment were identified.
Cores extracted from these sections revealed that 20% of Section 193-02
the cores were stripped, 4% of cores exhibited separa-
tion, and the rest 76% of the cores had no significant The road section is located in Cameron Parish of
damage. Four sections with moisture damage were inves- District 7. It was constructed in 1935 and carries an
tigated in detail. ADT of 2600. The section had a length of 14.3 miles
and the pavement structure consisted of a 10 in. AC
and 15 in. of a sand base on top of a fat clay subgrade.
Section 146-01 A microsurfacing maintenance was conducted in 2010.
The road section is located in Avoyelles Parish of District At the location of the core, the PCI was 85.9 and the
8. It was constructed in 1924 and carries an ADT of average PCI along the section was 90.7. The section
3500. The section had a length of 12.3 miles and the pave- receives about 62 in. of annual precipitation. However,
ment structure consisted of a 15 in. AC layer on top of a no significant water bodies were found near the section,
brown clay subgrade. A microsurfacing maintenance was see Figure 8c.
conducted in 2006. At the location of the core, the PCI
was 91.7 and the average PCI along the section was 91.8.
The section exhibits about 59 in. of annual rainfall and
Section 374-03
most part of the section passes alongside a stagnant The road section is located in Avoyelles Parish of District
waterbody within a 300 ft. proximity, which may have 8. It was constructed in 1928 and carries an ADT of 510.
significantly contributed to moisture damage underneath The section had a length of 18.4 miles and the pavement
the pavement, see Figure 8a. structure consisted of a 7 in. AC layer on top of a red
Bashar et al 11

Figure 8. Moisture-damaged sections with microsurfacing treatment.

silty sand subgrade. A microsurfacing maintenance was that microsurfacing-treated sections exhibited higher per-
conducted in 2003. At the location of the core, the PCI centage of moisture damage as compared with the
was 81.5 and the average PCI along the section was 88.4. untreated sections in all of the districts, a concern exists
The section receives about 57 in. of annual precipitation. that microsurfacing may contribute to moisture damage
Numerous small water bodies were found near the sec- in asphalt pavements. The reduction of water evaporat-
tion, as it is located very close to the Red River, see ing from the pavement surface due to the application of
Figure 8d. microsurfacing may be a contributing factor. Therefore,
The results of this analysis revealed that despite having an in-depth assessment of the effects of microsurfacing
a good-to-excellent PCI rating, these sections have signif- on moisture damage in asphalt pavements should be con-
icant moisture damage underneath the pavement surface. ducted, especially in areas with shallow GWT, consider-
Compared with the age of these pavements, microsurfa- ing factors such as precipitation, evaporation, and
cing treatments were applied only recently. Considering permeability of the pavement layers.
12 Transportation Research Record 00(0)

Conclusions and Recommendations state agencies to set guidelines for future maintenance
activities involving microsurfacing.
The study presented in this paper evaluated the effective-
ness of microsurfacing treatments in a setting where the
pavements are highly vulnerable to moisture damage due Acknowledgments
to heavy rainfall and the high GWT. The field perfor- The study was financially supported by the Louisiana
mance of pavements was modeled as a function of time Transportation Research Center (LTRC). The authors would
and the benefits were quantified in terms of rutting, like to acknowledge LTRC and Christophe Fillastre of
roughness, and overall condition of the pavements. LaDOTD for their assistance with data inquiry.
Based on the results of the analyses, notable conclusions
are summarized as follows: Author Contributions
The authors confirm the contribution to the paper as follows:
 Microsurfacing treatments were found to be most study conception and design: Mohammad Z. Bashar, Mostafa
effective in immediately improving the rutting A. Elseifi; data collection: Mohammad Z. Bashar, Momen R.
condition of the pavement in which it instantly Mousa; analysis and interpretation of results: Mohammad Z.
improved the RTI by about 18.1 units and slowed Bashar, Mostafa A. Elseifi; draft manuscript preparation:
the rutting deterioration by about 7.2 units/year. Mohammad Z. Bashar, Mostafa A. Elseifi, Zhongjie Zhang,
However, it also slowed down the deterioration of Kevin Gaspard. All authors reviewed the results and approved
roughness condition by 3.7 units/year. the final version of the manuscript.
 The long-term effectiveness of microsurfacing
treatments ranged from 10.0–11.3% for rutting References
and overall condition of the pavement. The effec-
1. Hein, D., C. Olidis, M. Darter, and H. Von Quintus.
tiveness results also indicated that the treatments Impact of Recent Technology Advancements on Pavement
are not beneficial if applied to a pavement with Life. Annual Conference of the Transportation Association
PCI, RTI, and RFI values greater than 85, 95, of Canada, 2003, pp. 1–24.
and 90, respectively. 2. Gransberg, D. D. NCHRP Synthesis 411: Microsurfacing.
 Microsurfacing treatment extended the service life Transportation Research Board, Washington, D.C., 2010.
of the pavements by 4.9–8.8 years on average. 3. Morian, D. A. Cost Benefit Analysis of Including Microsur-
Traffic and AC layer thickness were also found to facing in Pavement Treatment Strategies & Cycle Mainte-
be statistically significant in influencing the SLE nance. Pennsylvania Department of Transportation,
of the pavements. Harrisburg, PA, 2011.
 Microsurfacing was found to be cost effective. 4. Peshkin, D. G., T. E. Hoerner, and K. A. Zimmerman.
NCHRP Report 523: Optimal Timing of Pavement Preven-
However, applying the treatment with PCI . 95
tive Maintenance Treatment Applications. Transportation
will cause reductions in economic benefits.
Research Board of the National Academies, Washington,
 The optimum conditions for microsurfacing treat- D.C., 2004.
ments were estimated as 80\PCI  \85, 5. Broughton, B., S.-J. Lee, and Y.-J. Kim. 30 Years of
1500\ADT \3000, and 5" \AC thickness\7". Microsurfacing: A Review. ISRN Civil Engineering, Vol.
2012, 2012, pp. 1–7.
Considering the fact that microsurfacing-treated sections 6. Erwin, T., and S. Tighe. Safety Effect of Preventive Main-
exhibited higher percentage of moisture damage as com- tenance: A Case Study of Microsurfacing. Transportation
pared with the untreated sections in all of the districts, a Research Record: Journal of the Transportation Research
concern exists that microsurfacing may contribute to Board, 2008. 2044: 79–85.
moisture damage in asphalt pavements. Therefore, an in- 7. Labi, S., C. Author, and F. Pike. Long-Term Benefits of
Microsurfacing Applications in Indiana – Methodology
depth assessment of the effects of microsurfacing on
and Case Study. Presented at 85th Annual Meeting of the
moisture damage of the pavements should be conducted,
Transportation Research Board, Washington, D.C., 2006.
especially in areas with shallow GWT, considering fac- 8. Temple, W., S. Shah, H. Paul, and C. Abadie. Perfor-
tors such as precipitation, evaporation, and permeability mance of Louisiana’s Chip Seal and Microsurfacing Pro-
of the pavement layers. gram. Transportation Research Record: Journal of the
Other aspects of the treatment methods including Transportation Research Board, 2002. 1795: 3–16.
material characteristics of AC, aggregates, and applica- 9. Lyon, C., and B. Persaud. Safety Effects of Targeted Pro-
tion rates may also have significant effects on the perfor- gram to Improve Skid Resistance. Transportation Research
mance of the treatment activities, in addition to the Record: Journal of the Transportation Research Board,
factors considered in this study. In future studies, the 2009. 2068: 135–140.
effects of the methods of treatment application should be 10. Rajagopal, A. Effectiveness of Chip Sealing and Micro Sur-
investigated. The findings of this study can be used by facing on Pavement Serviceability and Life. The Ohio
Bashar et al 13

Department of Transportation and The U.S. Department Pavement Management System (PMS) of LADOTD.
of Transportation Federal Highway Administration, Louisiana Department of Transportation and Develop-
Columbus, OH, 2010. ment, Baton Rouge, LA, 2009.
11. Hixon, C. D., and D. A. Ooten. Nine Years of Microsurfa- 17. Lu, P., and D. Tolliver. Pavement Treatment Short-Term
cing in Oklahoma. Transportation Research Record: Jour- Effectiveness in IRI Change using Long-Term Pavement
nal of the Transportation Research Board, 1993. 1392: Program Data. Journal of Transportation Engineering, Vol.
13–19. 138, No. 11, 2012, pp. 1297–1302.
12. Hicks, R., J. Moulthrop, and J. Daleiden. Selecting a Pre- 18. Rajagopal, A. S., and K. P. George. Pavement Mainte-
ventive Maintenance Treatment for Flexible Pavements. nance Effectiveness. Transportation Research Record: Jour-
Transportation Research Record: Journal of the Transporta- nal of the Transportation Research Board, 1991. 1276:
tion Research Board, 1999. 1680: 1–12. 62–68.
13. Labi, S., G. Lamptey, and S.-H. Kong. Effectiveness of 19. Haider, S., and M. Dwaikat. Estimating Optimum Timing
Microsurfacing Treatments. Journal of Transportation for Preventive Maintenance Treatment to Mitigate Pave-
Engineering, Vol. 133, No. 5, 2007, pp. 298–307. ment Roughness. Transportation Research Record: Journal
14. Powell, B. Preservation Group Study Planned for NCAT of the Transportation Research Board, 2011. 2235: 43–53.
Test Track, Pavement Preservation Journal, 2012. 20. Khattak, M. J., G. Y. Baladi, and S. Ismail. Review of
15. Watson, D., and D. Jared. Georgia Department of Trans- Louisiana’s Pavement Management System: Phase I.
portation’s Experience with Microsurfacing. Transportation Transportation Research Record: Journal of the Transporta-
Research Record: Journal of the Transportation Research tion Board, 2008. 2084: 18–27.
Board, 1998. 1616: 42–46. 21. Powell, B. Preservation Group Study Planned for NCAT
16. Khattak, M. J., G. Y. Baladi, and X. Zhang. Development Test Track, Pavement Preservation Journal, 2012.
of Index-Based Pavement Performance Models for