WATER POLLUTION CONTROL RESEARCH SERIES * 11034 DUY
INVESTIGATION OF POROUS
PAVEMENTS FOR
URBAN RUNOFF CONTROL
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development, and demonstration
activities in the water research program of the Environmental
Protection Agency, through inhouse research and grants and
contracts with Federal, State, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, B.C. 20460.
-------
INVESTIGATION OF POROUS PAVEMENTS
FOR URBAN RUNOFF CONTROL
by
Edmund Thelen
Wilford C. Grover, PhD
Arnold J. Hoiberg, PhD
Thomas I. Haigh
The Franklin Institute Research Laboratories
Philadelphia, Pennsylvania 19103
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
Project #11034 DUY
Contract #14-12-924
March 1972
-------
EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or
recommendation for use.
-------
ABSTRACT
Laboratory and economic studies were undertaken to determine the
feasibility of utilizing porous pavements to alleviate combined
sewer overflow pollution and reduce the design parameters of storm'
sewer systems by allowing storm runoff to percolate back into the
ground.
Laboratory studies of candidate materials revealed a porous asphaltic
concrete containing 5.5% asphalt by weight and aggregate graded to
allow a water flow of 76" per hour to be the optimal porous road ma-
terial. Materials testing for stability, durability, and freeze-thaw
susceptibility proved this material suitable for use in road construc-
tion.
Asphalt Institute specifications were used to design roads with porous
asphaltic concrete surfaces and gravel bases for varying traffic
densities. Major design parameters considered were the load-bearing
capacity and permeability of the subgrade, expected maximum precip-
itation and depth of frost penetration.
Roads designed with porous asphaltic concrete, were found to be
generally more economical than conventional roads with storm sewers.
The economics of porous pavement were further enhanced by the added
value of benefits from combined sewer overflow pollution relief,
augmentation of municipal water supplies, improved traffic safety,
preservation of vegetation, relief of flash flooding and the aesthetic
and directional benefits of a colored porous surface.
This report was submitted in fulfillment of Project Number 11034 DUY,
Contract 14-12-924 under the sponsorship of the Office of Research and
Monitoring, Environmental Protection Agency.
iii
-------
-------
CONTENTS
Section
I Conclusions
II Recommendations
III Introduction
IV Porous Pavement Design
V Design of HotMix Asphaltic Concrete Pavements
VI Asphalt Pavement Structures
VII Materials and Methods
VIII Experimental Phase
IX Hydrological Considerations
X Drainage of Porous Asphalt Concrete Pavement
Structures
XI Pollution Effects of Urban Runoff
XII Bacterial Effects of Porous Pavement
XIII Special Pavement Types
XIV Conventional Pavement Costs
XV Maintenance and Resurfacing Costs
XVI Porous Pavement Benefits
XVII Market Demand for Porous Pavement
XVIII Additional Types of Porous Pavement
XIX Plans for Demonstration Pavement
XX Acknowledgements
XXI References
XXII Glossary
Paee
1
13
17
19
25
29
35
41
53
59
63
67
71
75
87
97
109
123
131
135
137
141
v
-------
FIGURES
Page
1 SUBGRADE SOIL PERMEABILITY REQUIREMENTS 33
2 GENERAL VIEW OF THE PERMEABILITY APPARATUS 37
3 GENERAL VIEW OF THE DURABILITY APPARATUS 39
4 PENETROMETER 40
5 WATER PASSING THROUGH A 4.5% ASPHALT CONTENT 44
CALIFORNIA SPECIFICATION SAMPLE
6 POROUS ASPHALT CONCRETE SAMPLES PRIOR TO FREEZE- 46
THAW TESTS
7 POROUS ASPHALT CONCRETE SAMPLES AFTER 265 CYCLES 47
OF FREEZE-THAW TESTS
8 POROUS ASPHALT CONCRETE SAMPLES AFTER 233 CYCLES 48
OF FREEZE-THAW TESTS
9 CUMULATIVE DISTRIBUTION OF MAXIMUM DAILY RAIN FALL 54
10 U.S. MAP - MAXIMUM DAILY PRECIPITATION ZONES 55
11 MEAN LENGTH OF FREEZE-FREE PERIOD 57
12 SUPPLEMENTARY DRAINS 61
13 WEB AND COVER PAVEMENT 124
14 ASSEMBLED PAVEMENT TEST APPARATUS 125
15 BRICKS WITH LUGS TO CONTROL SPACING BETWEEN 127
(SHADED AREA LOOSELY PACKED WITH GRAVEL)
16 BRICKS ASSEMBLED ON RODS WITH SPACERS BETWEEN 128
17 HEXAGONAL HONEYCOMB COVERED WITH PERFORATED RUBBER 129
SHEET LAID OVER GRAVEL
18 COVERED HEXAGONAL HONEYCOMB WITH SECOND PERFORATED 129
RUBBER SHEET SUBSTITUTED FOR GRAVEL
19 TESTS AT SITE . 134
vi
-------
TABLES
No.
1 Porous Pavement Benefits
2 Applications
3 Marshall Design Criteria
4 Requirements for Surface and Base Course Structure
5 Aggregate Gradations
6 Hot Mix Design Data
7 Freeze-Thaw Marshall Stability Values
8 Asphalt Penetration Values After Aging in the
Durability Apparatus
9 Bacterial Activity Data
10 Conventional Pavement Construction Costs
11 Drainage Facilities Costs, Including Installation,
Excavation, and Backfill
12 Average Costs of Conventional Pavement Construction
13 Service Life and Resurfacing Costs
14 Annual Maintenance Costs
15 Total Annual Maintenance and Resurfacing Costs
16 Economic Evaluation: Aesthetic and Ecological
Benefits '
17 Pavement Demand Estimates
18 Porous Pavement Design by Design Traffic Number
19 Equivalent Porous Pavement Design
20 Cost Comparison of Porous Pavement and Conventional
Pavement with Storm Drainage
21 Porous Pavement Benefits
7
20
29
31
41
43
49
51
68
84
85
86
88
91
95
106
113
114
115
116
117
-------
-------
SECTION I
CONCLUSIONS
Objectives and Benefits
The investigation of porous pavements was undertaken primarily because
of the potential of porous pavements for alleviating combined sewer
overflow pollution. Over-flow pollution is currently a problem for the
approximately 18% of the nation's population served by combined sewers.
During storms, interception sewer capabilities are too small to handle
the volume of flow generated, and the major portion of the stormwater
is outflowed directly into receiving watercourses. These stormwater
outflows carry with them as much as 25 to 40 percent of the year's
production of suspended solids, putrescible organic matter, and bacteria,
which tend to settle at the bottom of a sewer to be swept up and out
untreated to surrounding watercourses by the stormwater overflow. The
delivery of these untreated pollutants contributes significantly to the
pollution of surrounding waters. Porous pavement, by allowing stormwater
to percolate into the soil rather than overflow.combined sewer systems,
could alleviate much of this pollution.
Further, where separate storm sewer systems already exist or are to be
installed-, the use of porous pavement could produce substantial cost
savings. In the former case, the use of porous pavement could alleviate
the need to install additional capacity where the present storm-sewer
systems capacity is fully utilized. In the latter case, the use of
porous pavement would allow reduction in the design parameter of the
storm drainage collection system installed.
In addition to the primary objective of finding a means of eliminating
combined sewer pollution, as well as reducing the cost of storm drainage
collection systems, a number of other benefits were found to result from
porous pavement applications. These benefits are identified separately
below:
1. Storm-Water Retention. Evidence on the polluted character of storm-
water indicates that it may be necessary in the future to store such
water for subsequent treatment when system capacity is available. The
construction of porous pavement over an impervious membrane offers a
potential mechanism of storing polluted stormwater and slowly releasing
it for subsequent treatment.
2. Enhanced Water Supply. Substantial areas of the country are now
subject to water supply deficiencies. By allowing precipitation to
percolate back into the soil, the use of porous pavement could help to
alleviate water supply problems for the 24% of the nation's population
currently clustered in water shortage areas. This is particularly
important since existing water-transfer agreements will not be able to
supply the demand in many of these areas past the end of the century.
-------
3. Elimination of Curbing. Curbs and gutter could be eliminated
on low traffic density porous pavements, effecting considerable
cost economies as well as aesthetic enhancement.
4. Safety Improvement
a. Skid Resistance. Porous pavement overlays on conventional
surfaces have been found successful in preventing wet
skidding or hydroplaning accidents. For safety application
a 3/4" to 1" layer over normal dense pavement is used to
provide rapid lateral surface drainage. Such applications
have been used successfully on road surfaces in California,
Louisiana, Utah, and Pennsylvania and airport runways in
England and New Mexico.
b. Enhanced Visibility. Visibility of pavement markings is
expected to be improved because of rapid removal of water
and because of the marking material penetrating the voids
to present an oblique view. The enhanced visibility of
pavement markings would be an important factor in accident
mitigation during storms.
5. Use of Urban Debris. The porous pavements designed will require
a base reservoir capacity. There is the possibility that this
reservoir can be created using broken bricks, ceramic wastes,
solidified fly ash and other solid urban residue.
6. Low Maintenance Cost. As the recommended porous pavement
design consists of currently used road materials, maintenance costs
should not exceed the level of expense currently incurred.
7. Relief of flash flooding. About 300 square miles of new pave-
ment of all types is laid in the United States each year. Runoff
from such pavement may contribute to local floods downstream, which
can cause loss of life and property. Flash flooding presents an
ecological problem where storm water from large paved areas is drained
off directly into neighboring streams or other small water courses.
This practice produces considerable flash flooding and stream bed
erosion during periods of heavy rainfall. Porous pavement would pre-
vent this flash flooding and preserve local streams from erosion.
8. Preservation of Vegetation. Plants and vegetation along con-
ventional roads, particularly in areas of high pavement densities,
are often starved for water because the dry soil under the roads
tend to rob their supply. Porous pavement would restore natural
moisture to the benefit of roadside vegetation.
-------
9. Preservation of Natural Drainage Patterns - In contrast to
impervious surfaces, porous pavement would preserve natural drain-
age patterns. It is desirable to preserve natural drainage
patterns where paving is imposed on otherwise open areas whose
natural character is worth preserving and/or areas where the
surfacing will be only temporary..
10. Temperature and storm control. In dense urban areas a light
colored pavement would provide a cooling effect. Further, the
heated air from large expanses of dark asphalt paving is suspected
by meteorologists of causing thunderheads to develop on summer
afternoons. This may cause moisture-laden air to dump its water on
the cities where it loads the sewers, rather than carrying it over
to the farmlands beyond. A light-colored pavement should not have
this effect.
11. Color Infusion. A demand for colored porous surfaces was
evidenced in a wide variety of applications. Laboratory tests
indicated that colored roofing granules applied to a porous surface
offer a promising colored pavement at economical costs of 35£ to
60£ per square yard, but further field evaluation of the color's
durability are required. The use of naturally colored aggregates
would provide a satisfactory colored surface, but their use would
be economically limited to areas where colored aggregates occur
naturally. The use of light colored binders in place of an asphalt
binder was reported to be unsatisfactory because of a tendency of
dirt to adhere to the binders and because of evidence of low
durability.
12. The presence of puddles in parking lots and other areas
traversed by pedestrians, is not to be expected with porous pave-
ments .
Formulation and Test
A variety of conventional and unconventional materials were consid-
ered and the most promising ones tested to determine their physical
and economic feasibility as porous pavement materials. An open-
graded asphalt concrete was selected as the most suitable material
because of its superior physical characteristics, its low cost, and
its ability to be laid by conventional paving methods. Porous
portland cement surfaces were found unsuitable because of pavement
failure cause by the shifting and settling of the subgrade under
the load application point. Artificial turfs were found insuffic-
iently permeable for porous playground pavements. A resilient
porous surface of adhesively bonded chopped rubber, intended for
playground use, exhibited suitable physical characteristics, but
was judged too expensive for more than very limited use. Similarly,
pavements assembled at the site from factory-made components such as
bricks or honeycombs, were seen to be uneconomical.
-------
Porous asphalt concrete can be mixed in usual hot mix plants, and
compacted and laid with the machines customarily used for asphalt
concrete paving.
Three types of porous asphaltic concretes, corresponding to Asphalt
Institute, British, and California aggregate specifications, were
analyzed. Because of differences in size-of-aggregate specifications,
the infiltrations rates of these types varied from less than 5"/hr.
to more than 25"/hr. Each type was examined with asphalt binder
contents of from 4.0% to 5.5% of total weight.
The most porous of the open-graded asphalt concretes contained
aggregate graded in accordance with a California specification:
Sieve Opening (MM) Specification FIRL Product
1/2"
3/8"
#4
#8
#16
#200
12.7
9.51
4.76
2.38
1.19
.074
100
90-100
35-50
15-32
0-15
0-3
100
97
34
16
13
2
Five and one-half percent by weight of 85-100 penetration road
asphalt was the binder of choice. It is customary, when designing
a pavement, to test the asphalt-aggregate mix for its resistance to
stripping by water using ASTM D 1664. If the estimated coated area
is not above 95% in this test anti-stripping agents are added to the
asphalt.
Marshall stability tests were performed on all specimens to deter-
mine their load-bearing suitability in road use. All specimens con-
siderably exceeded the minimum Marshall stability criterion for
medium traffic uses.
Freeze-thaw tests were conducted to determine whether porous asphal-
tic concrete could withstand normal climatic cycles. Two samples
each of the Asphalt Institute, British and California specifications
were subjected to 265 freeze-thaw cycles. No physical dimensional
changes were noted for any of the samples after the test cycles, nor
was there any impairment of Marshall stability values or flow rates.
Durability tests were conducted to determine whether the heightened
exposure to air or water would produce excessive asphalt hardening
which would cause cracks to form in the road surface. Only the
California mix with 4.5% asphalt concrete exhibited excess hardening,
due mostly to its highly open structure. The California specifica-
tion with 5.5% asphalt content, however, proved to combine high
durability and high permeability, and was selected as the optimal
porous asphalt concrete surface.'
-------
Further tests were conducted to determine the effect of porous pave-
ment on the survival of aerobic soil bacteria. It was deemed
important that these aerobic bacteria flourish under porous pavement
to metabolize oils, animal and bird excrements, and other organics
that might otherwise tend to clog the system or pollute the water.
Rather sketchy tests for aerobic activity in soil located underneath
porous asphaltic concrete revealed no inhibition of these bacterial
processes.
Design of Porous Asphaltic Concrete Roadways
The design of porous asphalt concrete roadways equivalent to con-
ventionally constructed roads was found to depend primarily on the
load-bearing capacity of the subgrade, the expected traffic volume
and the reservoir capacity of surface and base. Specifications of
the Asphalt Institute were used to design the porous pavement road-
ways illustrated in the table below:
Requirements for Surface and Base Course
GBR
. DTN
2
2
2
2
2
2
2
1
10
20
50
100
1000
5000
Surface
Thickness
(IN)
4
4
4-1/2
5
5
6
7
Base
Thickness
(IN)
6
12
13
14
16
20
22
Reservoir Capacity
(inches of rainfall)
Surface Base Total
.60
.60
.66
.75
.75
.90
1.05
1.80
3.60
3.90
4.20
4.80
6.00
6,60
2.40
4.20
4.56
4.95
5.55
. 6.90
7.65
Because of the low load-bearing capacity of a wet subgrade a poor
subgrade (California Bearing Ratio = 2) was assumed in establishing
these designs. Traffic volumes are indicated by the Design Traffic
Number (DTN). Traffic volumes are designated by the DTN reading as
follows: 1-10, light traffic; 10-100 medium traffic, 100-5,000
heavy traffic (primarily highway.) The gravel base depths are
minimums provided by Asphalt Institute specifications. These
minimum base depths would have to be increased in designing porous
pavements for areas where the expected maximum precipitation
exceeds the indicated surface and base reservoir capacities. Thus
for an expected 5.4" maximum precipitation in one hour, typical of
Philadelphia, the minimum base thickness for all types of uses would
have to be 16". Adding additional base thickness is relatively
cheap and does not greatly affect the economic feasibility of porous
pavement.
-------
It was judged that a soil permeability of .042, sufficient to remove
a 5-inch rainfall in ten days, would prove adequate for most uses.
In areas with high rainfall in combination with soils of low water
infiltration rates, less than 0.02 inches per hour, consideration
should be given to either penetrating the low permeability soil
layer by sand or other type of drain if a higher permeability soil
horizon lies beneath, or by constructing nearby storage ponds to
retain the water until infiltration can occur. With the latter
eventuality, water penetrating the pavement layer would run off
laterally to be drained to the ponds.
Economic Evaluation of Porous Asphaltic Concrete Pavements
The economic analysis undertaken determined by pavement type the
cost of porous asphalt concrete in relation to conventional pave--
ments with storm sewers, the relative value and market for the
identified benefits of porous pavement, and the annual demand for
the pavement types considered. Cost estimates for porous asphaltic
concrete were based on Asphalt Institute specification assuming the
minimum base reservoir capacity of 16" that would be required in a
medium precipitation area like Philadelphia and assuming an average
winter frost penetration level. Where the total thickness of
surface and base exceeded conventional pavement thickness, an
allowance was made for extra excavation costs. In areas of higher
or lower precipitation, the cost of porous asphaltic concrete
construction would be correspondingly increased or decreased some-
what, because of changes in required base reservoir capacity.
Similarly, in areas of heavy frost susceptibility deep gravel bases
normally are required to prevent freeze-thaw heaving, and the cost
of porous construction would be correspondingly more favorable.
The results of the economic analysis are summarized in Table 1. In
most cases the cost of conventional pavement with storm sewers was
found to be higher than the cost of an equivalent porous asphaltic
concrete installation. Except for combined sewer pollution relief
and water supply enhancement, the benefits attributable to porous
surfaces are specific to the type of pavement and are discussed
separately by type of use.
The analysis of the economic benefit of combined sewer relief
involved consideration of a wide variety of corrective measures
covering a broad range of costs. Sewer separation was found to be
the most frequently adopted measure with costs averaging $10,200
per surface acre, but escalating rapidly in high density areas.
Separate road sewer systems have been constructed in Toronto and
other cities to relieve overflow pollution at an average cost of
$7,000 per surface acre in 1960 prices. Numerous less costly
measures - screening, chemical treatment, off-system storage and
sedimentation ponds - have also been undertaken at an average cost.
estimated by the America Public Works Association at $5,100 per
-------
LaJ
CO
CO
ID
O
O
Q.
c
o
l*~
c
It
o
o
o
^
CL
CL
ZJ
CO
0)
4-1
rtj
1_
1
to
"o co
E
O
CM
"Q "*x.
Q) -
T3 * '
-o
o
TD (1)
in C 4J
tn to O
O E 0 *
l_ H) 4- 5*9
C3 Q M~ -
0)
03
CM~"
"O >*
(U "^s.
~D X-N
i. 0) M- 3^
CU Q ^^-^
§J
nj ^ ^
> CM
"O "^s-
(D -CO-
TS *-^
~o
-O
tn ru O ^ *
O E O 5^
CJ Q lU*"'
ME
3 (D CM
P is ^
t- CO \
0 > Q >-
> ro -^
CD C -to-
CO .
3
r-i -D CM
3ro >
C E O
q co o
-> CM
CO CT\
0 O
1 1
Cn CM
CM n
0 0
-a- -a-
CM CM
LA LA
LA LA
CO CO
LA CO
VO \O
O CV-l
CM CM
r-. o
LA cn
CM CM
W> CM
r~-
0)
0)
4-1
to
cr
c cn
ro 01 *
4-1 to 0)
c a) -o
0) TJ i
-o c j=
* S cn
ui O
co -J in
o;
o
LA
1
O
O
^i
LA
V
CO CO
o o
1 I
CT\ CM
CM CA
O O
CM CM
-a- -a-
LA LA
LA LA
CO CO
CO vO
vo r-.
0 0
-a- CM
CO CO
1 VO
3" F^
LA f^.
CM -3-
LA
C
0)
E
CL
0
4J Q
to
c
to -
- -Q 4-1
01 3 .
ZJ tO CJ
CD
LA
I
0
LA
O
LA
V
CO CO
o o
J .
ro E i
O n a)
cc -t-i
O ro
>- > L.
4J 1
« o
0) (U
Q) C 4J 4->
C to (1J
> ro ^~ f L-
ro i -c o
3 i u o- c
-C O 3 01 O
01 3 o
1^ Ll_
O
CM
CM
O
O
_J
CO
-cr
vO
LA
LA
"3^
J|
"i
-o
laygroun
d-
-------
LU
"7 CO
§ fe
o
c
_a
3
o
O
c
o
4-1
ro
c
E
Hi
u-
o
S.5
J_l
4-i ra
(0 4-1
> O
L- Cl
0) ^
e; *^
a_
O
**j '.I-
ro
cu c
O
a>
Ol U-
(0
C JC
a
(0 - .
> OJ
*0 "*N.
o
-Q *
a
-a
*O fl)
tn c 4-»
t/> ro o - .
0 E fl) fi*
i_ d> M_ -^-^
O Q M-
0)
,_
ro ^-^
>
me*-1
w ro u*-^
o e <» ^?
i, m M- *-*
C3Q M-
^c
3
to
CNJ
*o >*
0) "-»
"O * *
^
^
T3 a)
C +-> '-^
ro o £?
01 M-
0 M-
*±
£
0)
ro
CL.
,,'
O
a)
a.
[_
O
O
^~
1
0
o
tn
V
o
o
CM
1
O
O
LTV
V
O
Ol
c
2
t_
ro
a.
O 0
3" -3"
CM CM
1 1
O 0
LTV Lf\
' '
Ln o
CO LA
o o
o o
1 1
0 O
LTV LTV
O O
irv
CM
1
IA O
V
4-1
H>
a)
L_
in
c
c en
ro 01
in
*-J in ol
c a) -a
a) "o i
TD 1 .C
3 en
Ul O
a> _i a;
c
^
O-
o
4-> ^~
01 01
01 >
t/>
c
i/i ro
in _o
a) i-
C 3 >-
yi 3
3 {/) c_>
OQ
O
O
*
1
o
LTV
O
LT>
V
d)
1
"o
a a) >
ro E i
o 3 a)
«: «
o ro
>. > i-
4J 1 01
C 3 T3
3 O O
0 -1 Z
O
a) a)
O C 4J W
c ro 01
>- ro ro i-
ro i -c o
3 i "- a. c
j: O 3 in O
01 3 O <
-------
acre. The economic value of porous pavement for eliminating over-
flow pollution was computed conservatively at the lower figure of
$5,100/acre, or an average of $5,54/y of pavement surface, based
on .a 19% pavement-to-surface area ratio. This economic benefit
would apply to the 18% of the nation currently served by combined
sewers.
By using porous pavement to divert stormwaters into the ground
rather than into a sewage system, the need for building treatment
plants can be mitigated and an additional substantial saving can be
realized over and above the savings reported above.
Water supply benefits are imputable to the 24% of the nation's pop-
ulation located in water deficit areas. The value of the additional
water was evaluated at 30£ to 60
-------
2 7
if it could be produced between 30£/y and 40£/y . Color infusion
was desired in 10% to 25% of the high design residential market,
depending on the level of urban-renewal activities, where the com-
bination of aesthetic and temperature control benefits would command
a premium of $l/y2. Porous pavement for vegetation preservation
would have only limited applicability in low-density suburban areas,
but would command an economic premium in a significant fraction of
the city residential street market. The elimination of curbing
would save between $1.50/y2 and $2.40/y2 in residential areas. This
benefit would affect most of the suburban residential market, but
only about half of the city market where safety consideration would
dictate retention of curbing on heavily travelled residential
streets.
Business Streets
Suburban and city businesses streets were evaluated separately
because of substantial differences in design levels. The relative
cost for porous and conventional design with storm sewers was
$6.38/y2 vs. 8.40/y2 and 7.66/y2 vs. 18.20/y2 for suburban and city
business streets respectively. A limited demand for color infusion
was evidenced for visually differentiating deceleration lanes, in
which use colored surfaces would command a $1.00/y2 to $1.50/y2
premium.
Country Roads
The relative cost of porous pavement was $6.44/y2 compared to
$9.20/y2 for conventional design with storm sewers. A very limited
market for colored surfaces was indicated, primarily for aesthetic
enhancement in scenic areas.
Highways
Substantial cost differentials were observed between porous
asphaltic concrete and conventional storm-sewered pavements. The
differential reflects primarily the extremely heavy expense of
installing storm drains in outlying areas where interceptor systems
are not available and stormwater must frequently be carried consid-
erable distances to a discharge point. The cost of porous design
was approximately half that of conventional installations for two-
lane highways and only a third of conventional costs for four-lane
highways. Considerable interest was expressed in the use of
colored surfaces for directional routing. Based on the cost of
current colored highway surfaces, the economic value of this use
would range from $2.50 to $3.50 per square yard.
10
-------
Playgrounds
The cos£ of porous pavement playgrounds was determined to average
$6.34/y in contrast to $4.55/y for conventional installation.
Again, part of the differential reflects the relatively high level
of the Asphalt Institute specifications in contrast to conventional
practice. Color infused surfaces were desired in most of the play
surface market with commandable premiums of $1.00/y to $2.20/y
depending on the type of installation.
General Economics
1. Except for very low-level design uses, the cost of porous pave-
ment is equal to or cheaper than the cost of conventional pavement
with storm or combined sewer facilities. For very high-level design
roads, the cost differential in favor of porous pavement is highly
significant.
2. Special ecological benefits of porous pavement - relief from
combined-sewer overflow pollution and water-supply augmentation -
have a potential for considerably enhancing the value of porous
pavement in a relatively large proportion (roughly 20%) of the
nation, by reducing the need for building more sewerage and treat-
ment plant capacity.
3. To the extent that the use of porous pavements over large up-
stream areas can mitigate floods due to runoff, the costs in lives
and property of such floods can be reduced.
4. While safety considerations would dictate the retention of
curbing in heavy traffic areas, the elimination of curbing in
suburban residential streets would effect a considerable saving of
from $1.50 to $2.40 per square yard of street surface.
5. Other special benefits of porous pavement, such as preservation
of vegetation, relief from flash flooding, and preservation of
natural drainage patterns, would command a significant economic
premium but would have a relatively limited market.
6. A broad demand for a color-infused asphaltic pavement was
evidenced for three primary uses: traffic control, aesthetic
enhancement and temperature control. A premium value of $0.30 to
$3.30 per square yard, depending on intended use, was indicated for
such a surface.
7. If a porous paving material were designed which is capable of
being laid during the winter, it would command a premium exceeding
$1 per square yard in the residential sector.
11
-------
8. The annual cost of maintenance and surface replacement must be
minimal in order for porous pavement to be a realistic alternative to
conventional surfaces. With few exceptions, these maintenance costs
were determined to average 2 percent or less than the cost of con-
ventional pavement with drainage. Although, theoretically, there
is an obvious economic tradeoff between initial cost and annual up-
keep costs, political realities preclude this tradeoff from being
effective both because of inconvenience to the public and difficul-
ties of obtaining substantial increases in maintenance appropriations.
9. Overall pavement demand for the major uses considered was
estimated at 967 million square yards annually. Of this total, the
highest users, in order of their respective contribution to total
demand, were low-design residential streets, county roads and
parking lots.
10. Given the significantly favorable economic differential height-
ened by a number of special benefits, porous surfaces have the long-
range potential for becoming the preferred alternative to many con-
ventional pavement uses. This potential is, however, dependent on a
convincing evaluation of the physical characteristics and maintenance
experience sufficient to prove that no problems exist which would
offset favorable economic indications.
12
-------
SECTION II
RECOMMENDATIONS
The physical feasibility of porous asphaltic concrete has been
demonstrated, in the laboratory, and its economic viability has been
demonstrated. Further research into porous surfaces is, however,
indicated in several areas. Recommended areas of further investiga-
tion are identified below:
1. Although the physical feasibility of porous pavement has been
demonstrated in the laboratory, porous pavement still needs to be
evaluated under actual service conditions. A demonstration program
should be undertaken to evaluate the performance of porous asphaltic
concrete over time under varying soil and climate conditions. The
demonstrations should be organized at sites- large enough to install
both a test porous pavement and a control conventional pavement for
comparison, and should be located in areas where it would be possible
to instrument both the conventional and test pavements. Low-Level
design surfacesresidential streets, parking lots, and playgrounds
are the most feasible surfaces for such a demonstration program.
The following determinations should be made in the demonstration ex-
periments :
a) The adequacy of porous pavement to imbibe water and avoid runoff
under varying levels of rainfall.
b) Its ability to survive weather, traffic loads, and other con-
ditions and how its performance changes with time and season,
particularly under freezing conditions and after application
of de-icing compounds.
c) The quality of imbibed waters, before and after percolation
into the soil.
d) The cleanability and maintainability of the porous surface.
If the porous surface is found to clog with debris, then
the efficiency, cost and machines for vacuum cleaning and
other maintenance procedures will have to be evaluated.
e) The effect of polluted water flowing through the pavement.
f) Public reaction to the pavement in terms of its comfort,
esthetics, quietness, etc.
g) Evidence of improved safety experiences on the porous
surface.
h) Any indication that fog could form, in unusual weather,
over the porous surface.
1.3
-------
i) The likelihood that there will be any time when the pavement
will be unable to accept water, and what then will happen,
for example when a rain falls after the pavement is covered
with ice.
2. The extent of the evidence uncovered on the polluted character
of stormwater per se indicates that serious consideration should be
given to means of retaining stormwater until there is sufficient
capacity in the system for it to flow to treatment plants. The use
of porous pavements installed over an impervious membrane offers a
promising method of retaining stormwaters until treatment capacity
is available. Laboratory experiments and field trials should be
undertaken to verify the feasibility of this use of porous pavement.
3. Since a very broad market demand was evidenced for a colored
porous pavement surface, the performance characteristics of roofing
granules adhered to porous surfaces should be fully evaluated. If
this material is found unsatisfactory, a research program should be
undertaken to develop a durable colored porous pavement surface.
4. Various kinds of solid urban debris have the potential for form-
ing a resevoir capacity under porous asphaltic concrete. Further
investigation of these materials is required to establish their
physical suitability and determine whether water can pass through them
without being polluted.
5. There are potential economic tradeoffs between reduced size of
collection systems and porous surfaces, such as using a porous surface
which can handle 3 inches/hour, of rain and a collection system to handle
the rare remainder. The economics of these tradeoffs need investigation.
6. A study remains to be made to elucidate the economic benefits of
using porous pavements as a potential means for reducing the capacity
of and flows to storm and combined sewer overflow treatment facilities
(to be required in the near future). The savings due to the reduced
cost of smaller treatment facilities and reduced operational and main-
tenance costs should be delineated.
7. The use of porous Portland cement concrete pavement of the type
tested \\rould not be feasible where there is shifting of the subsoil
and base. Further work would be required to-determine whether a
gravel course lightly bonded with asphalt would have the requisite
stability over an otherwise unstable soil.
8. A substantial demand was indicated for a durable, resilient, non-
abrasive porous surface for playground use, if such a surface could
be made available at a cost not exceeding $5 to $6 per square yard.
Further exploration of available materials should be undertaken to
determine whether such a porous surface can be produced in this price
range. Reclaimed rubber in particular, should be analyzed as a possible
material from which to achieve such a surface.
14
-------
9. There is some indication that snow or ice melts faster than normal
on porous pavement. Possibilities to enhance this characteristic
should be explored both to reduce the impact of snow and ice on
traffic flow and safety, and also to minimize the need for deicing
salts. Conversely the effects if any of deicing compounds on the
pavement structure and on buried structures (phone cables, water mains,
etc.) as well as the effects on soil bacteria and ground water quality
should be studied unless the use of salts can be dispensed with.
10. In some paving applications (bridge decks and advance base air
fields, for instance), it might be advantageous to prefabricate the
porous pavement, transport it as slabs, and drop it into place. As-
phalt concrete would not be suitable for this, and if required, other
types of porous pavement, though more expensive, should be developed
for such applications.
11. Simple how-to-do-it instructions should be prepared and tests
designated whereby porous pavements can be designed for given climate
and soil conditions. A manual for this purpose would show also "how
to predict the hydrological performances of a given design in a specific
situation.
Since friction coefficients, drainage slopes, inlets, drains, etc. may
not be the same for porous pavements as for conventional pavements, new
road and highway specifications for these purposes must be formulated.
12. Oil spills on a pavement are possible and their effects on a por-
ous pavement system should be determined under service conditions.
13. More extensive tests than those carried out in this project will
be required to prove conclusively that cleansing bacteria in the soil
are unaffected by the presence of porous pavement over them. Present
results are encouraging but not conclusive. These tests might best
be run on outdoor demonstration pavements subjected to polluted runoff
waters.
14. In view of the economic advantage of being able to lay a pavement
all year round, tests should be conducted to determine whether there
are any unusual problems in laying the porous asphalt concrete in winter.
15. Noises are generated by skidding of tires on pavement and the
rolling of tires on pavement; ambient noises are reflected or absorbed
by pavements. How porous pavements compare with normal ones in their
effect on noise levels is largely unknown, and, in view of the increas-
ing interest in noise pollution, should be measured. Tire squeal due
to skidding, and reflection of ambient noises should be reduced; roll-
ing friction noises levels from porous pavements need to be determined.
16. A considerable technology exists for the selection of asphalt
grades for specific applications, and for the improvement of asphalts
by additions of rubbers, resins, fillers, and fibres. The performances
15
-------
in service of open-graded asphalt concrete pavings should be
further surveyed and studied and guidelines prepared for selecting
the optimum asphalt grade and admixture, it required, for each
application.
It is likely that this formulation wave will produce an asphalt
concrete with considerably greater strength and resistance to
creep than the ones studied. If this can be achieved low design
pavements may not require a full four inches of opengraded asphalt
concrete and some economics may result.
17. When gravel is laid directly on a soil which is very soft
when wet, the gravel tends to sink into the soil, and become
clogged by it. To avoid this, a layer of sand or of filtering
fibres (such as a thick fibre glass felt is often placed between
the soil and the gravel. Research should be carried out to deter-
mine under what conditions such a layer would be advantageous in
the porous pavement system.
16
-------
SECTION III
'INTRODUCTION
A proposal dated August 13, 1969 was submitted to the Office of
Research and Monitoring Environmental Protection Agency, to develop
porous pavements; to determine their feasibility for a number of
applications and to build and test sample areas for demonstration
purposes. The proposal was accepted and work was initiated in Phase A
on October 30, 1970 by The Franklin Institute Research Laboratories.
The porous pavements were intended:
1) To absorb rain where it falls, preventing runoff and overflows
from combined sewers and treatment plants required to handle storm-
water loads.
2) To hold back stormwater surges to increase the capabilities of
existing facilities to handle them.
3) To allow the filtered water to percolate at its own rate into
the ground, and/or to store it for use;
4) To mitigate vehicular accidents and to minimize personal injuries
on playgrounds.
5) To avoid puddles in flat surfaces (such as parking lots) and
provide adequate drainage without crowns and vertical contouring.
6) To conserve water and land;
7) To present a satisfying appearance;
8) To minimize damage to the ecology, and
9) To provide a possible means of disposing of certain solid wastes.
This study was visualized in three phases. Phase A is a feasibility
study related to technological and cost capabilities of several types
of porous pavements. Phase A includes application requirements over
several types of soil, benefits to be achieved by avoiding runoff, and
acceptability, of the paving to governmental bodies, contractors, and
other interested groups.
This report covers results obtained under Phase A.
The bulk of the work consisted in studying and developing the most po-^
tentially useful type of porous pavement, and in determining the
economic feasibility of porous pavements in general and this one in
particular.
17
-------
The purpose of Phase B was to select and obtain test areas in
which the most promising designs for two or more applications
could be put to practical test.
Phase C was to establish a test regimen for instrumenting and
monitoring the demonstration areas to evaluate the performance
of the prototype porous, pavement and to compare it with con-
ventional pavement.
18
-------
SECTION IV
POROUS PAVEMENT DESIGN
1. Applications
Table 2 lists 10 representative pavement applications, and includes
the loads, speeds, and frequencies of traverse estimated for the
traffic on each. Also included are Special Considerations which
are discussed in more detail in the next section of this Discussion.
Estimated maximum loads are based upon the weight or pressure of
the vehicle or person, the size of its "footprints" and the spacing
between them. Looking at item 4 in Table 1, a tire inflated to 60
psi has a contact area of less than one square foot and might support
up to 2000 Ibs. of vehicle. However there is likely to be only one
wheel on a square yard of paving, hence the load per square yard also
is listed as 2000 Ibs.
The first three applications deal with human traffic and the next
two with passenger vehicle and light truck traffic. Three more in-
clude heavy truck loads and the last two deal with airport traffic
in terms of the biggest aircraft in operation, the C-5A.
The first five applications are subjected to relative low traffic
loads, hence'are more feasible than the others at sites where the
soil when wet may be quite poor in load bearing capacity.
Among the design features planned with porous pavement for many appli-
cations and locations are the following:
a) It shall carry the loads without damage.
b) It shall imbibe all or most of the rainfall on it, and
water from melted snow and ice.
c) It shall survive freeze-thaw and weathering; long life
is required.
d) It shall be hard to damage but easy to repair; maintenance
cost must be moderate.
e) It shall not plug up; removal of trash by occasional sweep-
ing or vacuuming may be permissible.
2. Water Flow Delay or Storage
A built-in reservoir capacity shall be designed to provide the follow-
ing:
19
-------
Table 2
APPLICATIONS
Speed Frequency
i High-Average Passages j Special
'ard mph per minute Consideration
* ^ \v
^- 03
E 3
3 CT
o Eui
CD
4-> X i-
-^
4J
O
03 03 CD O
E ^C d, M
4-< -
C
1
CD
o
CD
CD
O
co
o
LA
O
LA
CM
LA
CM
-a
c
3
o
s_
cn
>-
03
r
Q-
CA
O
"o
O
»v
(n
OJ
O
cn
c
Q.
CL
O
CO
LA
O
O
O
O
CM
O
O
0
CM
O
VO
4-J
O
_J
cn
c
^
03
-3-
4j
CD
CD
s_
4-J
CO
i
03
4->
C
CD
-a
in
CD
c£
LA
O
'o
O
CD
-o
in
cn
O
03
cn
4-J
03
o
O
CO
O
o
o
o
CM
,
O
-3-
O
._
<4-
14-
03
!_
4-J
-a
CD
4->
O
i_
4-1
in
CD
""
CD
4J
4->
_1
LA
LA
O
O
O
LA
O
0
O
1^,
O
-a
03
o
OL
CD
O
>
i_
CD
CO
VO
in
CD
a.
CO
o
14-
H-
03
s_
1-
o
o
m
o
o
o
o
o
o
o
LA
o
r^
4-J
CD
CD
i_
4-*
CO
in
in
CD
c
in
^
OQ
Q.
in
o
-a -a
^. 03
CO
4-" CD
CD 3
S u~
vo ca
CM
LA VO
VO
O O
0 O
O O
LA
O
O
0
r^
o
cr»
s~
c
^M
>-
03
2
cn
_
~i~
O
VO
o
o
o
CO
CM
o
o
CM
^
c
^»
^
4
^
c
o
I.
Q-
<
.
X
OJ
1-
C7)
C
c
OJ
Q-
O
u
T3
T3
C
OJ
-a
in
4-J
CD
ra
o
o
CA
0
o
O
o
VO
o
o
o
CO
CM
o
o
CM
-^
1
-^
cn
c.
»
"D
C
03
_J
4-J
L-
o
Q-
^
<^
x--\
r^-co o->
4-J
O
03
Q.
cn i-
C O
Q-
TD I/I
O C
03
O !-
C l-
LA
O I
z. o
CM
20
-------
a) Depending on the permeability and drainage of the underly-
ing soil, its percolation rate may be less than the rate
of rainfall; the pavement will store the incident'rain
and provide time for its slow percolation.
b) Similarly, the water collected within the pavement would
take some time to drain out, hence could be fed slowly to
a combined sewer avoiding present peak loadings, and/or
reducing the required size of a storm sewer.
c)
Alternatively, with an impervious membrane under the pave-
ment the water can be collected in the reservoir layer
from which it can be pumped, using a sump and pump, for
irrigation or other purposes.
3. Water Treatment
Two provisions for water treatment shall be incorporated:
a) The water will be filtered as it passes through the pavement,
the subbase and the subgrade.
b) Aerobic bacteria should be able to survive under the pave-
ment and help to destroy organic contamination that could
clog the soil or pollute the water.
4. Accident Mitigation
Safety features of the new pavement shall be developed:
a) By mitigating street -water from the pavement surface during
storms, the incidence of accidents due to hydroplaning and
loss of friction by vehicles will be greatly reduced.
b) By laying the pavement flat and eliminating gutters and
crowns, control of the vehicles, particularly at inter-
sections, will be improved. Since puddles will drain
into the pavement, it does not have to be crowned.
c) It may be feasible to impart to the pavement a surface
texture or configuration that will help to prevent skidding,
not only of vehicles but also of people on playgrounds and
sidewalks.
d) A light color will be incorporated when desired to improve
visibility and safety.
e) For playgrounds, a non-abrasive resilient surface will be
sought, which together with the elimination of puddles,
will reduce accidents.
21
-------
5. Appearance
From an esthetic point of view, this feature is of interest:
a) Elimination of gutters and crowns means the pavement can
be level with the parking area from a consideration of
runoff; this, together with lighter colors, should make
a more attractive road, particularly in residential areas
and parks.
b) Elimination of drainage ditches for unsewered roads should
make them safer and more attractive.
6. Ecology
In terms of ecology and environment, Mr. Richard James, Director
of the Schuylkill Valley Nature Center, raises these points:
a) Plants and vegetation along conventional roads often are
starved for water because the dry soil under the roads
tends to rob their supply. Water falling on most pave-
ments, if not collected in gutters, tends to cause erosion
of the bordering soil. These difficulties would not occur
with porous pavements.
b) The heated air from large expanses of dark asphalt paving
are suspected by metorologists of causing thunderheads
to develop on summer afternoons. This may 'cause moisture-
laden air to dump its water on the cities where it loads
the storm sewers, rather than carrying it over to the farm
lands beyond. A light-colored pavement should not have this
.effect.
7. Solid Waste Disposal
A useful goal, which was identified in a discussion with Mr. Rosenkranz
and is seen to be feasible, is as follows:
If a reservoir is to be created under the pavement, this could
be a place to deposit broken bricks, ceramic wastes, solidified
fly ash and other bothersome solid residues of the cities. However,
materials to be disposed of in this way will have to be known to
be non-polluting of the water passing through them.
8. Paving Design; Technical Constraints
All pavement designs in the end depend upon subgrade support to pre-
vent undue deformation and roughness or break-up of the pavement.
Additionally the pavement surface itself has to be durable to traffic
and weather. In the usual pavement design the pavement rests upon a
22
-------
base, and possibly a subbase which in turn is on compacted subgrade.
A base usually is required since only rarely does the compacted sub-
grade have sufficient supporting value and resistance to water and
frost conditions.
Rigid pavements, such as the Portland cement concrete type, depend in
design upon beam support, but at high stress locations such as at the
corner areas near joints they also can require localized subgrade and
base support. Flexible pavements, the asphalt binder type, depend
upon localized support, although with full depth design in which the
base also is bonded with asphalt, they more nearly approach the rigid
type in beam supporting value Bunder dynamic loading.
The usual pavement design prevents or attempts to prevent water from
entering the base and subgrade. Otherwise water seeping or flowing
through the pavement surface lowers the supporting value of most sub-
grades, and thus would require increased depth of the base structure.
If water is to penetrate through the base, the compacted subgrade and
the earth underneath, the problems will be to provide tlie required
drainage rate and/or water storage capacity; to obtain the necessary
minimum supporting bearing value of the subgrade; and to generate
sufficient frost resistance. Otherwise the pavement will have an un-
satisfactory economical life under its present and future expected
traffic.
The theoretical approach to porous pavement design thus requires that
a pavement have the required permeability to water, be durable in age
under traffic, be adequately supported by the base and subgrade under
conditions of water saturation and that the overall design be economi-
cal. An additional consideration is that particulate organic matter
flushed through the pavement voids be decomposed by bacterial action,
lest plugging of the subgrade occur.
9. Factor Fabricated Paving Structures
Two general categories of porous pavements were considered:
a) Those fabricated in a factory and delivered to the site for
assembly and installation, and
b) Those delivered as bulk materials to the site and there
spread and compacted into place.
Factory-fabricated structures can be precisely shaped and sophisti-
cated in design, and materials can be specified within relatively
narrow limits.
Four examples of these were studied:
The Portland cement concrete structures, special brick and honeycomb
structures, Astroturf rugs, and various products studied for play-
grounds as described in Section XIII.
23
-------
The one bulk material that was found suitable is open-graded as-
phalt concrete. When its feasibility became evident, the factory-
fabricated materials, by comparison, were seen to be uneconomical.
All of the factory fabricated pavements have higher materials costs
than open-graded asphalt concrete. They require labor at the site
to assemble and install them, whereas the asphalt concrete can be
mixed in existing paving plants and laid with a Barber Greene type
machine and rolled with standard equipment.
Factory fabricated porous pavements might have utility for emergency
or military airstrips, and for roa'ds that have to be installed in a
minimum of time. They could also be useful for bridge decks or other
structures not laid on soil.
In view of the very minor impact of such limited applications on run-
off and sewage problems, they were not investigated further.
24
-------
SECTION V
DESIGN OF HOT-MIX ASPHALTIC CEMENT PAVEMENTS
In terms of cost and practicality, the porous pavement should be made
from conventional paving materials, and applied at the site by conven-
tional pavement-laying devices. An open-graded hot mix asphalt concrete,
alone of all the materials considered, meets these requirements.
Several specifications of open graded asphalt concrete pavements have
been promulgated by various agencies; generally these pavements are
used either to avoid wet skidding or hydroplaning accidents, or as a
revetment, to protect dam faces, or other embankments. (Ref. 2).
For the safety application, the material is laid 3/4" to 1" thick over
a normal dense pavement. Water drains laterally through this course, to
its edges. Applications of this sort have been used successfully for
25 years in California, which attests to their mechanical and photo-
oxidative durability (Refs. 1 and 6).
A dense concrete runway was overlaid with an open-graded asphalt concrete
course on a runway at the British Royal Air Force Farnborough Airport in
England. It successfully withstood multiple landings of aircraft with
12,500 pounds force wheels loadings, and has survived several years of
weathering (Refs. 3 and 5). It is claimed that open-graded asphalt con-
crete pavement provides better protection against wet skidding and
hydroplaning than is afforded by grooving the pavement.
Other sgencies trying out asphalt concrete porous pavement for safety
reasons are the U.S. Air Force at Albuquerque, New Mexico, the States of
Louisiana, Colorado and Pennsylvania, etc. The widespread interest in
this material should cause it to be commercially available also for
water conservation purposes.
Asphaltic base pavements, primarily prepared with asphaltic cement
binders can be made with a variety of asphaltic products, as cutbacks
and emulsions. This investigation, however, was concerned only with
the use of asphaltic cement binders since until the cutbacks and emul-
sions become thoroughly set, the water stripping tendency would be great
in porous asphaltic pavements. As defined herein porous asphaltic con-
crete is a graded aggregate cemented together by asphalt cement into a
coherent mass, with sufficient interconnected voids to show a high
permeability to liquid water. [The maximum intensity of rainfall ex-
pected in the United -States for a duration of one hour from a storm
that might occur once in 50 years is 5 inches per hour (Pennsylvania =
2.2 to 2.8 inches/hr). For a 10 year period it is 4 inches per hour
(Pennsylvania = 1.6-2.2 inches/hr).] In manufacture, the graded aggre-
gate is dried by heating to 275°-300°F, and mixed with hot asphalt
cement. The hot-mix material is hauled to the paving site, where it is
applied in a spreading machine in smooth layers, and compacted to design
density by heavy rollers.
25
-------
Properties required in porous asphaltic concrete pavement are:
It should be durable and not deteriorate under weather or
traffic.
It should be economical and not employ expensive materials.
It should be safe under traffic. As will be discussed later,
porous asphaltic concrete has exceptional skid resistance pro-
perties .
The following factors are primary in design of porous asphaltic con-
crete pavement:
1) Aggregate Gradation
The gradation required to obtain a porous asphaltic concrete pave-
ment is of the "open" graded type as contrasted to the "dense" graded
type which is capable of close packing.
2) Type and Quality of Aggregate
The aggregates selected for porous pavement construction should meet
the requirements of the standard specification for "Crushed Stone,
Crushed Slag, and Crushed Gravel for Bituminous Macadam base and Sur-
face Courses of Pavements", ASTM C-693-71, with one exception and
one addition. The exception is in requiring the gradation to be of
the open graded type of the reasons described below. The addition '
should be a requirement for a soundness test as specified in ASTM
D-694-62, "Crushed Stone, Crushed Slag, and Crushed Gravel for Dry-
Bound or Water-Bound Macadam Base Courses." This is reouired to
determine if the aggregate is susceptible to disintegration by water.
3) Asphalt Cement Grade and Content in Mix
Possible factors in open graded mixes because of their relatively
high permeability to air and water are their durability to freeze-
thaw and to asphalt film oxidation. To offset this, experience has
shown that a minimum asphalt content of 5.5% by weight in the mix
with a relatively soft asphalt can provide a very satisfactory service
life. Suitable asphalts have penetrations of 85 to 100 dmm, ASTM D-5.
A requirement commonly specified for mixes and of special importance
in water permeable mixes is set forth in "Coating and Stripping of
Bitumen-Aggregate Mixtures", ASTM D 1664. The asphalt coated aggre-
gate should retain an estimated 95 percent of the coated area after
water immersion of the mix for 16 to 18 hours at 77°F. (25°C).
Mix and pavement thickness design methods for conventional pavements
are quite standardized and are given by publications of the Asphalt
Institute. Porous pavement design differs in that the so-termed
26
-------
open aggregate gradation is selected as opposed to the usual'
dense gradation. The desired gradation as determined by this
investigation is discussed under the section, "Surface Course."
The design requires that the surface course have the required
stability to traffic to not be displaced by traffic loading,
either by rutting or raveling or shoving.
Overall thickness design of porous asphaltic concrete pavements
is determined by a number of factors, viz:
1) Subgrade supporting value in a wet compacted condition. This
determines the depth required of both the base, sub-base, and
surface course. The California Bearing Ratio (CBR) technique
commonly is used to measure the load bearing characteristics
of the subgrade soil. The test consists of measuring the load
required to cause a plunger of standard size to penetrate a
soil specimen at a specified rate. The CBR is the load in pounds
per square inch, required to force a piston into the soil a cer-
tain depth, expressed as a percentage of the load in pounds per
square inch, required to force the piston the same depth into a
standard sample of crushed stone. Usually depths of 0.1 inch or
0.2 inch are used. Penetation loads for the crushed stone have
been standardized. The resulting ratio of bearing values is known
as the California Bearing Ratio.
2) Water storage capacity. Subgrades, that is the existing soil
structure, vary greatly in their capacity to infiltrate water.
With soils showing low infiltration rates, the base thickness
may need to be increased to have sufficient storage capacity to
permit the rainfall to infiltrate in a reasonable time, herein
arbitrarily selected as 10 days. With some soils, as clays and
fine silts, the infiltration rate may be so low that other means
may be required to introduce the water into the earth, as infil-
tration pounds or sand drains. Such soils, however, do not make
suitable subgrades and usually it is more economical to replace
them with selected borrow with better support value. Such re-
placement ,also would have the advantage of having a higher storage
capacity and infiltration rate.
3) Frost penetration. If the climate is such that frost will penetrate
deeper than the anticipated thickness of surface and base and the
subgrade is susceptible to frost heaving, additional subbase will
be required. Frost 'penetration in inches has been related to an air
freezing index. Charts of these freeze indices and frost penetra-
tions are available in standard publications as, Asphalt Institute
"Drainage of Asphalt Pavement Structures", MS-15 (1966).
On an average, for one year in ten, values for frost penetration in
the eastern United States vary from 12 inches in Northern Tennessee
to around 100 inches in middle Maine. Acceptable soils to resist
frost heaving have been defined by a triangular chart, Asphalt
27
-------
Institute publication MS-15 (1966), with essentially the clay and
silt contents of the subgrade to total less than 40% by weight.
4) Design Traffic Number. An asphalt pavement structure deflects
downward under a moving load, and returns almost to its original
position after the load has passed. This "almost" is important,
since under continued traffic the permanent deflections add to
altered pavement profiles, and may possibly result in pavement
cracking. By taking into account the anticipated traffic the
base thickness may be increased to reduce the deflection of
the subgrade under traffic load to an acceptable value.
The anticipated traffic is calculated as a Design Traffic Number
(DNT), which is the average number of equivalent 18,000 pound
single axle loads per day expected in the heaviest traffic lane
during the design period (normally 20 years). The method of cal-
culating the DNT is given in detail in the Asphalt Institute Manual
Series No. 1, "Thickness Design," the classifications of traffic
are as follows:
DTN
1-10
10-100
100-10,000
Traffic Classification
Light (Parking lots, residential streets)
Medium (City business streets)
Heavy (Highways)
28
-------
SECTION VI
ASPHALT PAVEMENT STRUCTURES
a. Surface Course
Asphalt paving mixes may be designed and produced from a wide range
of aggregate blends, each suited to specific uses. The aggregate
composition may vary from course to fine particles. To sort out
the differences in grading, porous asphalt paving mixes have been
classified into three types according to their water infiltration
rates, 1) less than 5"/hr, 2) 5"/tir to 25"/hr and 3) greater than
*25"/hr. The Asphalt Institute, British and California gradations
shown in the experimental phase of this report represent types 1,
2 and 3 respectively. The permeabilities and seepage rates also
vary considerably with changes in asphalt content.
The Marshall test procedures for designing paving mixtures give the
guidelines for preparing a mix capable of withstanding different
traffic conditions. The criteria are in Table 3.
Traffic Category
No. of Compaction
Blows (each end of
specimen)
Table 3
MARSHALL DESIGN CRITERIA
Heavy Medium
75 50
Light
50
Test Property Min. Max. Min. , Max. Min. Max.
Stability (ibs) 750 - 500 500
Flow (.01 inch) 8 16 . 8 18 8 20
% Air voids 3 5 3 5 3 5
As the experimental data of this program will show, stability and
flow requirements obtained were more than adequate for each of'the
three types of our classification. The % Air voids, however, which
are mentioned in the Marshall Criteria are not appropriate for a
porous surface course. The Asphalt Institute gradation gave values
as high as 8.2% air voids but provided a seepage rate of only 1.15
in/hr. This would be adequate only for minimal rainfall conditions.
Seepage rates generally required can be adequately provided by both
the British and California aggregate gradations specification for a
porous concrete at asphalt concentration of 4.0% for the British and
4.0 to 5.5 for the California. For purposes of road paving durability
to prevent too rapid a hardening of the asphalt it is desirable to
have the highest asphalt content possible in the mix. Too much asphalt
would separate out under traffic and hence maximum asphalt content would
be limited by this factor. 5.5% asphalt content with the California
aggregate specification would appear to be the optimum.
29
-------
If we take the Asphalt Institute's recommendation of a 4" minimum
surface course we obtain a 0.60" rainfall reservoir capacity within
the surface course with the 5.5% California specification and a
seepage rate of 176 in./hr. This was the maximum obtainable. The
combination of the surface course with the base course enhance total
rainfall hold capacity as will be shown later.
b. Base Course
Asphalt pavement structures must be designed and built to support
the heaviest traffic volumes and loads for a particular application.
By increasing the depth of the base course, loads are spread conically
over large areas, thus reducing the loading intensity until the sub-
grade will support the load without undue deformation.
A porous asphalt pavement structure must not only carry the loads
without damage but must also:
1. Imbibe all or most of the rainfall on it, and water
from melted snow.
2. Survive freeze-thaw and weathering.
The base course depth can and will provide for the load and act as
a water reservoir.
The base course generally consists of crushed gravel. It is used
under the surface course to distribute loads, store water and prevent
freeze-thaw problems. In place of gravel, broken bricks, ceramic
shards or other urban solid wastes or suitable strength and insolu-
bility might be disposed of here. For relatively uniform-sized lumps,
a void volume of 38 to 46% might be expected. If a 40% void volume
is assumed the average, a twelve-inch layer would hold 4.8 inches of
rainfall.
c. Subgrade
Water continually in contact with silty or clay types of subgrade
soils will saturate and decrease the strength of the subgrade and,
in turn, lower the load-carrying capability of the roadway. Therefore
an ideal soil for use as a subgrade must have adequate load-bearing
capacity wet and dry. This ideal soil must be well drained that is,
the water table is so far below the surface that the rise of water by
capillary action will not come close to the surface. It must be
permeable to water, so water is transmitted relatively quickly down
through it. Finally, it must be quite porous so it has the capacity
to inbibe large quantities of water in its surface layers.
A likely soil to meet these requirements would be a sandy silt some
distance above the water table. The coefficient of permeability for
such a soil must be greater than .01 ft/day if a full reservoir is
to be emptied within a reasonable amount to time. The particle
30
-------
for this type soil ranges from sand to clay, but the clay content
is low enough to leave good void volume and permit capillary flow
downward at a suitable rate. If highly sandy a low content of
asphalt often is added to improve interlocking of the sand particles
and improve load bearing properties of the soil; good transmission
5 characteristics will remain satisfactory. If it is too clayey, the
soil compacts very well and carries the load when dry, but porosity
is lacking and transmission of water through the clay is very. slow.
Clays may also absorb more than enough water to fill their voids,
thereby swelling and becoming very low in load support value.
It is possible to treat a high clay soil with lime or cement to
aggregate the grains, making it more porous and less likely to be
swollen by water. Such soil treatments have been developed over the
years, and can be considered when the top layer of soil is deficient.
If the underlying soil lacks permeability, however, treatment of the
top layer would not correct the problem.
d. The Porous Asphalt Pavement Structure
In designing the total structure for use as a porous pavement with-
out the use of membranes or drains it is necessary to ascertain the
following:
1. Design traffic number (DTN).
2. Load bearing capacity of the wet subgrade (CBR) and its
permeability.
3. Permeability, strength, flow, and durability of the
surface course.
The Asphalt Institute recommends that a minimum of 4 inch surface
thickness be used regardless of the DTN. Using this as a basis-,
Table 4 was compiled for a low load bearing ratio and varying DTN's
Table 4 V
REQUIREMENTS FOR SURFACE AND BASE COURSE STRUCTURE
CBR
2
2
2
2
2
2
2
DTN
1
10
20
50
100
1000
5000
Th ickness
(in.)
4
4
4-1/2
5
5
6
7
Surface
Reserve! r
Capaci ty
(in. Ra infal 1 )
.60
.60
.66
.75
.75
.90
1.05
Base
Th i ckness
(in.)
6
12
13
14
16
20
22
Base
Reserve i r
Capaci ty
(in. Ra infal
2.40
4.80
5.20
5.60
6.40
8.00
8.88
1)
31
-------
The reservoir capacity for the surface and base courses are based
on 15 and 30% air voids respectively. These are conservative.
If we assume that the load bearing capacity of the subgrade is
poor (CBR = 2) when wet and we wish to design a pavement structure
for a residential street (DTN = 10) in an area where we might expect
a 5" rainfall in a 24 hour period, we can see from the Table that
a 4 inch surface course and a 12 inch base course would be adequate
to handle both the traffic and rainfall. On the other hand, if the
DTN is one and the rainfall expected is still 5"/day, then we must
still use the 4" and 12" surface and base course thicknesses.
Once the reservoir is filled it must be emptied by percolation through
the subgrade. Figure 1 shows the permeability of subgrade soil re-
quired to empty the reservoir within a reasonable amount of time. If,
for example, we wish to remove a 5" rainfall from the reservoir in ten
days, the subgrade soil must have a permeability value of at least
.042 ft/day. This value would be typical of a fine sand and silty soil.
32
-------
UJ
a
<
cc
o
CD
0.10
0.01
0.001
IDEAL CONDITIONS (FROST MINOR)
0.10
1.0 10.0
TIME TO DRAIN BASE RESERVOIR (DAYS)
100
Figure 1. Subgrade Soil Permeability Requirements
33
-------
-------
SECTION VII
MATERIALS AND METHODS PHASE
The materials necessary to construct a porous asphalt concrete top
surface course consist of (1) appropriate penetration grade asphalt
and (2) properly sized crushed aggregate. It was decided early in
the program to use 85/100 penetration grade asphalt (ASTM procedure
D-5) as a standard since it is in wide use today on the nation's
highways. This material has shown good strength and durability
properties over the years of its use. The crushed aggregate in our
experiments was comprised of two different aggregates, a coarse and
fine graded crushed limestone and sand.
The hot-mix asphalt paving mix is a combination of crushed aggregates
uniformly mixed and coated with asphalt cement. The porosity of this
mix is dependent upon the gradation of the aggregate portion, the amount
of asphalt used as the binder and the degree of compaction. The aggre-
gates must be dried and the asphalt cement be sufficiently fluid for
proper mixing and workability. This necessitates that both the aggre-
gate and asphalt be heated sufficiently prior to mixing. It is also
required that heating be continued or that as little heat as possible
be lost during mixing and compacting.
a. Marshall Method
A procedure in wide use today for both laboratory design and field
control of the asphaltic load bearing properties of hot-mix paving
is the Marshall Method. It is applicable to hot-mix paving mixtures
using penetration grades of asphalt cement containing maximum par-
ticle sizes of one inch or less in the aggregates. This method has
been standardized by the American Society for Testing and Materials,
as given by ASTM designation D-1559, "Resistance to Plastic Flow of
Bituminous Mixtures Using Marshall Apparatus." The procedure con-
sists basically of preparing hot mix asphalt concrete samples in 4"
dia. by 2-1/2" high molds, maintaining the proper temperatures to
keep aggregate dry and the asphalt fluid enough to be mixed uniformly
in a mechanical mixer. The uniform mix is compacted in the molds
with a 10 Ib hammer being dropped a distance of 18" on first one side
of the specimen and then on the other; 35, 50, or 75 blows can be
applied to each side depending on the traffic category for the mix.
This procedure was used throughout the course of the project with
certain exceptions as presently will be noted.
b. Permeability
Since the program was concerned with porous asphalt mixes a perme-
ability test was developed based on the. Marshall specimens. By
this procedure Marshall samples could be tested prior to being ex-
tracted from their molds. A compacted specimen while still in the
mold is bonded tightly at the walls of the retaining structure and
thus wall leakage problems are eliminated in the permeability
35
-------
apparatus. The ASTM designation D-2434 method for measuring the
permeability of compacted soil was modified and utilized for this
work. The general arrangement of this apparatus is shown in Figure 2.
A constant head water bath supplies water flow through the specimen,
after it has been evacuated with an aspirator for 15 minutes. The
hydraulic gradient across the specimen is measured with manometer
openings placed directly above and below the specimen. Test runs
were conducted by varying the head pressure in 1/16" increments to
establish a region of laminar flow through the specimen. This con-
dition is necessary for application of Darcy's law which measures
permeability. Basically, it states that the velocity of a liquid
through the specimen is directly proportional to the hydraulic
gradient across it. From this relation it is possible to compute
th'e coefficient of permeability (K in ft/day) for the specimen as
follows:
K
_
Ath
where
K
Q
L
A
t
h
Coefficient of permeability (ft/day)
Quantity of water discharged (Cu ft)
Distance between manometers (specimen height) (ft)
Cross-sectional area of specimen (Sq ft)
Total time of discharge (days)
Difference in head on manometers (ft)
c. Freeze-Thaw
The permeability apparatus also was used to aspirate samples
prepared for freeze-thaw tests. The 4" dia. by 2-1/2" high samples
were made using the Marshall Method. These samples were drilled half-
way through, to their centers, for housing a thermocouple well con-
sisting of a 1/4" diameter by 1/4" long stainless steel sheath attached
to a bakelite head. This well prevented erroneous readings of air tem-
perature which might be read while the sample was in the freezer. A
thermocouple readout system was used for measuring the sample's
temperature. The following freeze-thaw cycle was used:
1. Aspirate (5 minutes)
2. Soak (15 minutes)
3. Drain (15 minutes)
4. Freeze (30 minutes) time required to bring center of
specimen to 0°F
5. Thaw (30 minutes) time required to bring center of
specimen to 50°F
6. Repeat
Control specimens (no water soaking) were also run.
36
-------
^fra^&i^%"^" ^Lv&'iw*^^;^!^^^***w^'*i 'fiE^N? ; *;U^*I«T*^V>&*;;
;,«ifV- '-^''*"-"4*-,H ,-- ---t4" ;!«»?; '': US :, P-,5 'IS
Figure 2. General View of the Permeability Apparatus
37
-------
d. Durability
Since the porous asphalt concrete will allow sufficient quantities
of air and water to pass through it was necessary to measure the
durability to air oxidation of such a mix. A durability apparatus
was constructed to house samples still in their Marshall compaction
molds as shown in Figure 3. Air heated with electrical heating ele-
ments and controlled by a thermocouple to 140°F +5°F was passed into
the manifold chamber housing which can hold up to 12, 4" diameter
x 2-1/2" length samples. The pressure head was maintained at 5" water
head through the use of the bleeder valve in the end of the chamber.
Samples exposed in this apparatus at various time intervals were re-
moved and the asphalt extracted for penetration measurement using ASTM
procedure D-5. The penetrometer is shown in Figure 4. The ASTM desig-
nation D-1856 procedure is used for the extraction process. Blank
determinations also were run to make comparisons with aged specimens.
e.Tank Specimens
For covering the Surface of the biological test tank, 40" x 52" in area,
asphaltic concrete blocks were prepared in one foot square molds con-
structed for this purpose. The mix procedure utilizes the Marshall
Method but a hydraulic press was used in place of the standard com-
paction hammer for the compaction process. The two and one-half inch
thick by one square foot samples were prepared in this manner and were
laid over a prepared soil bed for the biological and chemical test
program to be carried out in the test tank.
38
-------
4-1
td
n)
&,
§
N
3
0)
,a
4-1
0)
C
a)
O
60
H
39
-------
\
Figure 4. Penetrometer
40
-------
SECTION VIII
EXPERIMENTAL PHASE
a. Aggregate Gradation
During the course of this program it was discovered that three
aggregate gradations previously had been established for producing
a porous asphalt concrete mix. The maximum aggregate particle sizes
for all three did not exceed one inch and asphaltic mixes containing
them hence could readily be tested by the Marshall Method. Their
specifications (see Ref. 1, 2, 3) and the results of our (FIRL) blend
were as follows:
Table 5
AGGREGATE GRADATIONS
% Passing Thru Sieve
Sieve
1"
3/4"
1/2"
3/8"
#4
#8
#16
#30
#100
#200
Open i ng
(mm)
25.4
19.0
12.7
9-51
4.76
2.38
1.19
.595
.149
.074
Asphalt
1 nsti tute
100
95-100
85-95
44-56
13-22
3-8
1-4
FIRL
100
98
93
47
21
3
3
British
100
90-100
60-75
25-35
12-18
8-13
3-5
FIRL
100
92
76
26
19
12
3
Ca 1 i forn i a
100
90-100
35-50
15-32
0-15
0-3
FIRL
100
97
34
16
13
Coarse and fine washed, crushed aggregates, were blended with sand to
produce the above FIRL specifications using ASTM method designation
C-136 for sieve analysis. ASTM procedures C-127 and C-128 were used
to measure the bulk specific gravities of the aggregates which were
necessary for determining the % air voids of the prepared samples.
By definition , coarse aggregate is retained on a #8 sieve whereas
fine aggregate goes thru it.
b. Marshall Values
The Marshall method provided the basis for determining the optimum
asphalt content to produce a surface course with suitable strength
for use on a roadway designed for medium traffic. This calls for a
50 blow compaction to each side of a prepared sample in a compaction
mold 4" in diameter and 2-1/2" in length. We used a temperature of
325°F to mix our aggregates with the 85/100 penetration grade asphalt.
A minimum mix stability value desired for such a roadway is 500 Ibs,
which was greatly exceeded in all mixes. Three samples of each
41
-------
aggregate gradation were prepared at three different asphalt con-
tents for tests of stability, flow, permeability and percent voids.
The stability and flow values as shown by data of Table 6 were
obtained after the samples had been measured for permeability. The
Marshall method requires that samples be immersed in a water bath
held to 140°F+ 1.8°F for 30 to 40 minutes before the stability and
flow tests. We used an Instron test machine to compression-load the
specimen at a constant rate of 2 inches per minute deformation to the
point of failure. The total number of pounds required to produce
failure at 140°F is recorded as its Marshall stability value. The
deformation recorded for the specimen at its maximum strength is
expressed in units of 1/100 inch as its flow e.g. if the specimen
deformed 0.10 inch the flow value is 10. A flow value of from 8 to
18 is desired for medium traffic pavements and was satisfied by all
specimens.
c. Permeability
The coefficient of permeability values (K) shown in Table 6 were ob-
tained after 24 hours of aging at room temperature. As stated pre-
viously the samples were not extracted from the molds during this time
since the bond between the specimen and the mold wall provided a seal
to prevent side leakage errors for the permeability measurements. The
molds were an integral part of the permeability apparatus. The constant
head water bath was used to supply water flow through the specimen after
it was evacuated with an aspirator for approximately 15 minutes. The
hydraulic gradient across the specimen was measured with manometers
placed directly above and below the specimen. Test runs were conducted
at heads increasing by 1/32 in. for the California gradation, 1/6 in.
for the British gradation and 1/4 in. for the Asphalt Institute gradation.
These different head increments were required for each of the three grada-
tions in order to establish the region of laminar flow where the velocity
of water through the specimen is directly proportional to the hydraulic
gradient across the specimen. This requirement must be maintained for
Darcy's Law to be used in the computation of the coefficient of per-
meability. Departures from the linear relation (turbulent flow) became
apparent at a greater than one inch head for the Asphalt Institute grada-
tion and varied from 1/16 in. to 1/2 in. for both the British and Cali-
fornia gradations. The laminar flow region was difficult to determine
for the California and British gradations due to the fast rates through
such open structures which accounts for the wide range of K values ob-
tained for these specimens. Figure 5 is a view of water passing through
a California specimen. It is easily seen from this Figure that high
permeability values were obtained with this mix.
The data in Table 6 correlated extremely well for specific gravity and
percent air voids. The permeability values, however, when compared
to percent air voids do not in most cases correlate. It is felt that
the turbulent region of flow through such open structures can not be
eliminated from interfering with measurements of K values, and hence
42
-------
Table 6
HOT MIX DESIGN DATA
A-!
A-2
A~3
Avg
A- 4
A-5
A-6
Avg
B-1
B-2
B-3
Avg
B-4
B-5
B-6
Avg
B-7
B-8
B-9
Avg
C-1
C-2
C-3
Avg
C-4
C-5
C-6
Avg
C-7
C-8
C-9 -
Avg
C-10
C-ll
C-12
4.0
4.0
4.0
5.0
5.0
5.0
4.0
4.0
4.0
4.5
4.5
4.5
5.0
5.0
5.0
4.0
4.0
4.0
4.5
4.5
4.5
5.0
5.0
5.0
5.5
5.5
5.5
Gradat ion
Asph. Inst.
Asph. Inst.
Asph. Inst.
Asph. Inst.
Asph. Inst.
Asph. Inst.
Bri t ish
British
British
British
British
British
Bri tish
British
British
Calif.
Calif.
Cal if.
Calif.
Calif.
Cal If.
Calif.
Calif.
Calif.
Calif.
Calif.
Calif.
SpG
2,183
2,246
2,195
2,208
2,264
2,240
2,237
T72T7
2,185
2,142
2,228
2,185
2,166
2,186
2,156
2,169
2,227
2,338
2,219
17I5T
2,018
2,012
2,021
2,017
1,956
2,062
1 ,950
1 ,
-------
Figure 5. Water Passing Through a 4.5% Asphalt Content
California Specification Sample
44
-------
accounts for the discrepancy. The values obtained, however, for
permeability are felt to be the most reliable indication of the
porosity of each specimen. The asphalt Institute mixes and the
5% asphalt content British mix are not adequate for handling heavy
rainfalls. This is shown by their very low seepage rates. All
other mixes are more than adequate to handle excessively large
rainfalls i.e. greater than 15 in./hr of rainfall.
d. Freeze-Thaw
Two samples each of the Asphalt Institute, British and California
gradation specifications at 4.5% asphalt content were used for
freeze-thaw tests. They were prepared as discussed previously and
put through the aforementioned freeze-thaw cycle. The condition of
these specimens before and after 265 cycle is shown in Figure 6 and
7. Samples numbered one of each set were used as controls i.e. put
through the cooling cycle, but not water soaked. Samples numbered
two of each set were put through the regular freeze-thaw cycle. Two
samples of each of a 5.5% Asphalt content (California mix) and a 4.5%
asphalt content (California mix) containing 2% latex by weight of
asphalt were also tested. Figure 8 shows the condition of these
specimens after 233 cycles.
There was no physical dimensional change noted for any of the samples
after the test cycles. All the samples were tested for changes in
Marshall Stability and flow values. The results of this test are
shown in Table 7, together with the range of values obtained for sam-
ples not put through the freeze thaw cycle. Values had not previously
been obtained, however, for the latex samples. The data shows that
both the stability and flow values for the cycled samples were not
affected, and hence heaving problems would not occur for a drained
porous pavement surface course.
e. Durability
Two samples each of the Asphalt Institute and British gradations
with 4.5% asphalt content were tested for asphalt penetration
values after two weeks in the durability apparatus. The asphalt
was extracted from the aggregate using ASTM specification D-1856
and measured for penetration using ASTM method D-5. The penetration
values obtained are based on the depth a standard needle can pene-
trate into a sample maintained at 77°F. Conditions in the durability
apparatus were 140°F at a 5 in. pressure head. The results as shown
in Table 6 showed little if any difference when compared to the blank
value obtained. The conditions then were changed to 150°F + 5°F at
3 in. pressure head and an eight week continuous duration. The
apparatus was operated at a 3 in. pressure head since 10 samples were
being run continuously and all were readily porous which prevented
obtaining the previous 5 in value. Two samples each of 4.5% asphalt
content California, British and Asphalt Institute gradations were run as were
45
-------
ASPHAIT INST. (Samolfi #1\
ASPHALT INST, (Sample
BRITISH fSamnlf
BRITISH (Sample
CALIFORNIA (Sample.«}
CALIFORNIA (Sample K)
Figure 6. Porous Asphalt Concrete Samples
Prior to Freeze-Thaw Tests
46
-------
ASPHALT IMST, {samole #1)
ASPHALT !NST« (sample #2)
BRITISH (sample
BRITISH (sample
CALIFORNIA (sample #1}
CftUFORHiA (sample K)
Figure 7. Porous Asphalt Concrete Samples
After 265 Cycles Freeze-Thaw Tests
47
-------
CALIFORNIA (Sample 11)
5.5X Asphalt Content
based on total weight
CALIFORNIA (Sample #2}
5.5% Asphalt Content
based on total weight
CALIFORNIA (Sample jR)
4.52 Asphalt Content
based on total weight
(2% latex content based
on weight of Asphalt
CALIFORNIA (Sample #2)
4.51 Asphalt Content
based on total weight
(2% latex content based
on weight of Asphalt
Figure 8. Porous Asphalt Concrete Samples
After 233 Cycles Freeze-Thaw Tests
48
-------
GO
UJ
__i
1
1 1
( |
CQ
00
)
i
ro
-o h-
CD (D >-
ro O CD
!- >. CO l/l
CD U -D
> C -Q
<£ * ro v '
4-J
co
3 o
o o
U- ~\
i in
. n
JD
ro > '
CO
1
(D
N in
0) CD
CD
s_ o
U- >
O
o 3
-C
O 1
O
~
ro
ro
s_
C3
Q.
l/l
oXO
Q)
D
E
ro
CO
O , CA O
+!+(
-a- o LA o o
CO LA CNl LA
CO r , LA i CA
+1 +1
CNl PA CA CO
o o o o o
O CNl O CNl CA
CNI o vo -a-
LA LA LA LA LA
vO vO vO NO vD
CNl CNl CNl CNl CNl
4-> 4-> ro
cc c
in in O
^ , . , 14
Q. Q-
1/31/1 s- i- ro
4->
£= C C
O -O O
O O O
*. - V ' '
CN! CNl i-
-------
two containing 5.5% asphalt content California gradation and two
with 4.5% asphalt California containing 2% latex by weight of asphalt,,
The asphalt penetration values for these samples are also shown in
Table 8 along with the asphalt penetration value obtained for a 4.5%
asphalt content California gradation which was aged for 8 weeks at room
temperature.
Clark (see ref. 4) determined that one week of weathering of a mix in
an oven at 150°F was equivalent to one year of natural weathering.
This was based on service tests over a four year period. Using this
criterion ,the results in Table 8 show the hardening process of the
asphalt which may be expected after an eight year period in the field.
The California (most porous) at 5.5% asphalt content is equivalent to
the less porous Asphalt Institute and British mixes at the 4.5% asphalt
contents. The 4.5% California mix shows the most hardening effect for
the asphalt, which of course is due to its highly open structure. The
addition of latex to this mix apparently had no effect. The 4..5%
asphalt content California mix at room temperature also shows increased
hardening. Since this was aged without heating or air flow through it,
its hardening is attributed to the sorption of its light oils into the
aggregate. This process is speeded by raising the temperature, and is
believed to account for most of the hardening shown in Table 8. Appar-
ently the 5.5% asphalt content would have to be the minimum amount for
use with the California gradation. This information has been borne out
by tests being conducted by the California highway transportation agency.
f. Traffic Hazard Considerations
Considered here are pavement improvements over and above the desira-
bility of water infiltration into the soil.
Skid Resistance And Visibility
Research on factors improving the friction characteristics of pavements
have shown that braking forces on wet pavements are more dependent on
the surface characteristics of pavements than on the tire or tread (See
Ref. 5). Additionally, this study found that aircraft tire wear was
not excessive on open graded mixes and that braking forces were not
directional as they are with grooved pavements. The basic finding was
that drainage under the footprint was sufficient to prevent aqua-planning
and would not necessitate the use of a grooved tread, with the attendant
disadvantage of inadequate drainage when the tire ribs are worn away.
A similar finding has been reported from California with open-graded
mixes permitting lateral drainage of rainfall to the edge of pavements
(See Ref. 6).
A secondary, but nevertheless important feature in accident prevention
is the better visibility of pavement markings expected during
50
-------
Table 8
ASPHALT PENETRATION VALUES AFTER
AGING IN THE DURABILITY APPARATUS
(Asphalt penetration blank = 94)
Asphalt Penetration Value, dmm.
Sample
Al
A2
Bl
B2
Cl
C2
Dl
D2
Ll*
L2**
% Asphalt
4.5
4.5
4.5
4.5
4.5
4.5
5.5
5.5
4.5
4.5
Gradation
Asphalt Institute
Asphalt Institute
British
British
Cal i fornia
Cal i fornia
Cal i fornia
Cal ifornia
Cal i forni a
Cal i forni a
2 weeks @
140°F, 5" head
92
92
92
92
8 weeks @
150°F, 3" head
28
30
33
26
17
18
30
29
16
43
(room temp.)
* % Asphalt based on total weight.
** Contained 2% latex based on weight of asphalt.
51
-------
periods of heavy rainfall with porous pavements. Water cannot build
above the surface to change the reflective appearance.
These two features, improved skid resistance and visibility, are
considered very important plus factors in pavement design, espe-
cially since these properties can be obtained at little or no
additional pavement surface cost.
52
-------
SECTION IX
HYDROLOGICAL CONSIDERATIONS
Hydrological considerations are important to the concept of porous
pavement in two respects: first, the expected maximum precipita-
tion will affect the required sub-pavement reservoir capacity as well
as in some instances, the viability of porous pavement itself and
second, the extent of the frost-period will dictate variations in the
thickness of conventional pavement, thereby influencing the relative
economics of conventional versus porous design. These two areas are
discussed separately below.
1. Maximum Expected Precipitation
Data on recorded maximum daily precipitation in the nation is presented
in Figure 9, "Cumulative Distribution of Maximum Daily Rainfall," (data
based on maximum recorded U.S. point rainfall at 288 first order sta-
tions), and Figure 10, U.S. Map of Maximum Daily Precipitation Zones.
The expected maximum daily precipitation is the prime determinant of
the level of pavement reservoir capacity necessary to store runoff be-
fore it percolates into the soil. It is anticipated that the following
gravel'layers will be necessary to provide adequate reservoir capacities:
7 inches in areas of 3" maximum 24 hour precipitation, 14 inches for 6"
maximum daily precipitation and 28 inches for areas of 12" anticipated
maximum daily precipitation.
As shown in Figure 9, the distribution of precipitation amounts is as
follows: 3 inches or less maximum daily precipitation - 15%, 3 to 6
inch maximum - 36%, 6 to 9 inch maximum - 36% and 9 to 12 inches maxi-
mum daily precipitation - 12%. Geographically, (Fig. 10) most of the
western half of the United States and the Great Lakes region are in
generalized areas of less than 6" recorded 24 hour precipitation. Most
of the South, Middle Atlantic and New England regions are in areas of
6 to 10 inch maximum 24 hour precipitation, whereas most of the southern
coastal region has greater than 10 inch expected daily maximum precipita-
tion. Although the high level of precipitation in these coastal areas
might lead to the assumption that the areas are not suitable for porous
pavement, the high porosity and correspondingly high percolation rate
of southern coastal soils is probably sufficient to compensate for the
heavy level of expected maximum precipitation.
2. Freeze/Thaw Susceptibility
The length of the frost period is a major determinant of pavement
susceptibility to freeze-thaw damage. In areas of extensive frost,
ice forms in frost-susceptible soils causing a raising of
53
-------
t
CO <
\
\
D <
v
\
\
'
3-
\
\
\
\j c
\
"\
D
\
\
\
cc
30 <
b
<
Q
Q
£!
a«
X
UJ v
o
5 r-
=> 0
X*
<
2 ;
-i!
Uj m C
1 algi
I i^c
1 2
I
\
\
\
D <
? fe
a> K
o u-
H- u
»f
< U.
o . O
O -^
: uj S^
' Q. to
o ^ u h-
" o^ 1- <
- ID X
o ro ts g
II 211 5 y
. uj o
: ' 3 ^
2 uj?
J §zi
t ^2
d Ss
C X <
L t- ce
\
\
\
3- <
_1
2
z
<
a
to
^
Q
a:
o
a
or
V5
z.
o
s
CO
ff
LU
Q
OL
0
\
\
\J C
81
T
CO
cri
-------
2
X
2
o:
O
i
CM
Z
O
<
51
a
tr
CL
Q
2
cc
§
tr
en
LU
o
z
CO
LU
X
0
z
CO
Al
H
CO
LU
X
o
z
g
Al
a
CO
LU
X
0
CO
V
D
u
UJ UJ [2
§5 »
s >
o cc -
o uj PO
u. <" Ci
OK z
(- y cc
Q. £ UJ
Si*
cri 3:
= ^^
uJ
-------
the pavement surface and resulting in seasonal frost-heave and
settlement. This condition is a major cause of pavement surface
deterioration.
Clean sands and gravel under pavements do not exhibit detrimental
heave and subsidence from seasonal freezing and thawing. Conse-
quently, in areas of freeze-thaw susceptibility, additional base
material, above that required for pavement support, is used to
overcome road surface freeze-thaw damage.
The relative duration of the frost season is indicated by the mean
length of the freeze-free period as illustrated in the zonal map
attached CFig. 11). In areas of frequent frost, corresponding
approximately to the 120 and 150 freeze-free day zones, a total
surface and base thickness of 28 to 36 inches is recommended to
provide protection from freeze-thaw damage. In these areas, the
base layer required to prevent freeze thaw damage equals or ex-
ceeds the gravel base reservoir capacity required for porous
pavement.
In contrast, a total high design base and surface thickness over
normal soils of 16 to 18 inches is generally required .in areas
of limited frost period, such as California. In these areas ad-
ditional base and excavation costs may be incurred, particularly
in low-design roads, to create adequate water storage capacity.
56
-------
o
H
fl)
a)
cu
4-1
ff
13
cu
60
H
ooe
57
-------
-------
SECTION X
DRAINAGE OF POROUS ASPHALT CONCRETE PAVEMENT STRUCTURES
If tests indicate that the subgrade is too impermeable to drain
the base course reservoir within a reasonable amount of time, it
will be advisable to provide a supplementary drainage system.
There are a number of popular drainage designs capable of ad-
equately removing the water remaining in the base course.
Several of these are:
1. French Drain
This system utilizes coarse open-graded rock in relatively deep
pits or trenches. This type of drain can be expanded out at the
lateral edges of a roadway to provide a deeper system with more
water holding capacity, so that the water has more time to per-
colate through a less permeable subgrade. Such as system is
shown in Figure 12a.
2. Sand Drain
This system is similar to the French Drain but the coarse open-
graded rock also contains enough fines to prevent intrusion of
adjacent soil which may tend to clog the drain. This clogging is
dependent upon the nature of the subgrade soil. Again this system
could be expanded to provide more water holding capacity (see
Fig. 12b).
3. Two-Layers Systems
With this type system a subbase is provided as a filter medium.
The coarse open rock drain layer (base) is protected by a suitable
filter layer (subbase) such as clean concrete sand. This system
generally provides excellent resistance to clogging and provides
excellent drainage capacity after numerous years of service. This
system is shown in Figure 12c.
4. V-Trench Water Removal to Pond
A positive method of removal of water contained in a French Drain
type system is to use the base material as a drain for transport
to a relatively shallow V-trench at a low point in the cross sec-
tion of the roadway. If a heavy volume of water is expected, it
may be advisable to obtain greater drain capacity by the construc-
tion of a two-layer system. If an appreciable gradient is involved,
a cross-drain should be placed at the downhill end of the cut.
This will intercept any water flowing longitudinally which if not
drained could saturate the fills and cause slumping of the fill
slopes. The V-trench shown in Figure 12d could be emptied into a
59
-------
storage pond or other suitable drainage system. The profile of a road
showing cross drains which empty into the V-trench is shown in Figure
12e.
5. Pipe Drains
Subgrade soils which are for all practical purposes impervious to water
would necessitate the use of pipe drains. The pipe is generally per-
forated with 1/4 to 3/8 inch diameter holes, placed in two or more
double rows 90° to 120° apart running lengthwise of the pipe. Materi-
als of construction include aluminum alloys, asbestos cement, bitumi-
nize fiber, cast iron, clay, concrete, and plastic. They are available
in sizes ranging from 4 inches to 120 inches and larger in diameter.
The sizes and type pipe to be used would be determined by the condi-
tions set for drainage. Such a system is shown in Figure 12f.
60
-------
I
V
d t>
t>
o
A (
/
A 4 P>
i "^
A
EXPANDED
FRENCH
DRAIN
'A:-v:
/
IliiB
SUBGRADE
ounrMi,c.
/
$:'&
;.;&.;;
A./.'/
BASE
EXPANDED
SAND DRAIN
(OPEN ROCK
CONTAINING
FINES )
(a)
(b)
SURFACE
-BASE
SUBGRADE
(c)
FILTER
BASE
SURFACE
SUBGRADE
SURFACE
SUBGRADE
BASE-g£
SURFACE
BASE SUBGRADE
DRAIN PIPE TO
V-TRENCH OR
OTHER DRAINAGE
SYSTEM
(f)
Figure 12. Supplementary Drains
61
DRAIN
PIPE
-------
-------
SECTION XI
THE POLLUTION EFFECT OF URBAN RUNOFF
A literature investigation was carried out primarily to determine the
water pollution hazard from urban runoff. What actually is present in
urban runoff that would constitute a potential pollution hazard? Where
does it come from? How and why does it reach the receiving stream?
What can be done to prevent it? And how would a porous pavement system
alleviate the above problem?
In 1969 The American Public Works Association under contract to The
Federal Water Pollution Control Administration (Contract No. Wa 66-23)
made a study on the causes and remedies of water pollution from surface
drainage in metropolitan Chicago, 111.(?) The study points out that
street refuse transported by runoff is a significant source of pollu-
tion in the receiving stream.
Here street refuse is defined as the accumulation of materials on the
street, sidewalk, or along the curb and gutter which can be removed by
sweeping. Components or particles larger than 1/8" diameter are called
litter; those smaller than 1/8" diameter are classed as dust or dirt.
In May 1967 a survey was made of 500 cities of various sizes and loca-
tions in the nation. The purpose of the survey was to identify the
major refuse producing sources in the various communities, une rrunareu
forty-nine cities replied. The survey disclosed six major sources of
street litter:
(1) spillage from overloaded trucks, (2) yard refuse (leaves and lawn
clippings), (3) improperly used trash receptacles, (4) debris from
construction and demolition, (5) roadside dumping, (6) poor public
cooperation.
The amount of refuse deposited from various sources was determined by
.a studyC7) of 18 representative areas in Chicago to be from 0.5 to 8
pounds/100 ft. of curb/day. The average varied from 4.7 for commercial
areas to 2.4 for single family residential areas. A total yearly
summary of estimated street litter and dust from a 10 acre residential
area in Chicago showed the following components: rags, .018 tons;
paper, .432 tons; dust and dirt, 6.6 tons; vegetation 2.22 tons; inor-
ganics, 1.08 tons. The most significant component in terms of poten-
tial pollution hazard is the dust and dirt fraction which varied from
0.4 to 5.2 Ibs./lOO ft. of curb/day. Of this material approximately
3% was soluble and readily transportable. The soluble material con-
tained appreciable amount of contaminants measured as: BOD (5mg/g),
COD (40mg/g), nitrogen forms (0.48 mg/g), phosphates (>0.05 mg/g),
total bacterial count (>107/g), coliforms (>10°/g), and fecal enter-
ococci (5400/g).
63
-------
It is seen that dust and dirt are the predominant pollutants.
The coarse or crude litter aside from representing an esthetic
pollutional hazard, can be washed into street inlets and catch
basins where they decompose and create an additional oxygen
demand on the receiving waters.
The study disclosed that the liquid remaining in catch basins
between precipitations tends to become septic and the solids
trapped in the basin take on the characteristics of anerobic
sludge. During even minor rainfall the displacement can re-
lease a major amount of this septic material into the sewer
system.
On the basis that contact between street litter and precipitation
runoff water could occur with a 14 day accumulation of material and
that all of the soluble and transportable BOD would be discharged
into the street inlets during a two hour storm, the peak flush effect
on receiving waters per mile of street could be 160% of the raw
sewage BOD and 800% of the secondary treatment effluent during the
two hour runoff period.
Thus one can see that "clean" storm waters are polluted. Rain
scavenges air pollutants out of the atmosphere, flows across
roofs across grass sprayed with insecticides and herbicides and
fertilized with inorganic nitrogen and phosphorous, across pet and
bird droppings along street gutters and finally through catch basins
where the flow may displace perhaps 2 cubic yards of stagnant water
and carry with it some of the digested solids from the bottom of the
catch basin. By the time the storm water reaches the interceptor,
it may exceed the polluting strength of sanitary sewage. When salts
from snow and ice control, lead and other contaminants from automobile
exhausts are added, the storm water may have a wide range of undesirable
characteristics.
Overflows from combined sewers are identified as sources of pollution.
In dry weather the small volumes of domestic sewage in 'large combined
sewers results in low velocities. Solids therefore settle out along
the sewer line. Storm flows tend to scour this material out and carry
it to the overflow. It ,has been estimated that from 3 to 5% of the
total organic load reaching the sewer leaves by overflow.
The American Public Works Association was asked to estimate the cost
of separating combined sewers nationwide. They estimated a cost of
$48 billion dollars (1967). They also estimated a cost of $15 billion
for alternate methods of treatment and control.
64
-------
Another finding of the survey indicated that less than 20% of the com-
bined sewer overflow regulators could be adjusted to meet various flow
criteria. Of the 10,025 regulators found in the jourisdictions inter-
viewed, 40% were nothing more than simple wires, many with design fea-
tures which are not responsive to overflow regulation.
One question remains to be answered. How would a porous pavement sys-
tem alleviate or at least reduce the above problem?
Depending on the pore size of the system in question, much of the dust
and dirt fraction would be washed through to the underlying soil. The
soluble portion of the dust and dirt fraction along with any-.other
biodegradable wastes would be washed through to nature's own waste
treatment plant, the soil.
Pesticides, herbicides, air pollutants, nitrate and phosphate ferti-
lizers would also be washed through where they would probably be
metabolized by soil bacteria.
Catch basins would not be needed with a porous pavement system, thereby
eliminating the septic sludge which builds up in them and saving the
cost of having them cleaned periodically.
The only problem remaining is that of the debris which is too coarse
to be washed through the pavement system. This problem can only be
solved by better "housekeeping" practices on the part of the area res-
idents and more efficient street cleaning procedures in our municipal-
ities .
65
-------
-------
SECTION XII
BACTERIOLOGICAL EFFECTS OF POROUS PAVEMENTS
The activity of soil bacteria serves to decrease the quantity of or-
ganic pollutants carried into the soil by surface waters. Under normal
conditions in the top layers of soil, the bacterial processes by which
the pollutants are oxidized draw on oxygen dissolved in the surface
water or soil moisture. If air is excluded from the process, the dis-
solved oxygen is rapidly depleted and other sources of oxygen are used.
The first alternate source is the oxygen in nitrates and nitrites con-
tained in the soil, and the results of the reduction of these materials
is the appearance of ammonia nitrogen in the percolating waters. The
second source of oxygen is from the reduction of sulfates, with sul-
fides, usually as iron sulfide accompanied by hydrogen sulfide, as an
end product.
While the degree to which pollutants are oxidized before reaching the
water table is dependent on the time the pollutant takes to traverse
the area of bacterial action, reducing the air supply to this area
would lead to reduction of the alternate sources of oxygen and generate
objectionable odors and compounds. Over a longer period the alternate
sources of oxygen could be exhausted and little change would occur in
the dissolved pollutants.
Experimental
In determining the effect of porous pavement on the activity of soil
bacteria, two test tanks were used. One tank contained no pavement
while the other contained a porous asphalt concrete pavement sealed to
the tank walls. Water was allowed to percolate through the test tank
at five day intervals and samples of the percolated water were drawn
off.
The tests performed were:
1. BOD - Used to determine whether pollutants were percolating
through the soil in the tank and to assess bacteriological
activity. Unless pollutants are added to the tanks, a pro-
gressive lowering of the BOD should occur since the amount of
soil, hence amount of pollutant, in the tank is finite. If
BOD tests are carried out without seeding, and if the BOD
disappears from the tank, it would indicate that bacterial
food had been exhausted. Intentional addition of a bacterial
food should reestablish bacterial activity and a BOD should
be obtained without seeding the samples.
2. COD - Used as a check on BOD. COD should exceed the BOD and
should normally decrease if BOD decreases.
67
-------
3. Ammonium Nitrogen - Used as first indication of the
establishment of anaerobic process.
4. Sulfides - Used to indicate a well established anaero-
bic process.
Test Results
The test results are tabulated in Table 9. BOD results varied
but were generally at a low value during the first three sampling
intervals. In order to boost, the BOD, the tanks were treated
xvith Peptone Solution between the 3rd and 4th sampling periods.
The tank without the test pavement immediately showed a BOD
increase but the tank with the pavement showed a continued de-
creased BOD. The 5th period BOD samples showed an algal growth
since the dilutions were inadvertently exposed to light. This
would account for the lack of BOD since oxygen was being re-
plenished by the Algae. The lag of BOD recovery in the tank con-
taining the test pavement seems to indicate that the solution of
organic "food" was not transmitted to the lower layers of the
soil in the tanks as rapidly as in the tank with no pavement.
This would be advantageous since the material would not penetrate
to the water table as rapidly and bacterial processes would have
longer to act on it.
The BOD results are supported by the COD results. The COD de-
creased by the 4th sampling period but did not increase there-
after as rapidly for the tank containing the pavement as for the
tank containing no pavement.
In none of the samples was ammonium nitrogen increased above the
background value, and sulfide was not detected. This would indi-
cate that the bacterial processes in both tanks remained aerobic.
Table 9
BACTERIAL ACTIVITY DATA
Sample No.
With Without With Without With Without With Without With Without
'vo 4.86 4.86 1.62 12.42 <\,o 'UJ
5 day BOD 4.86 7.56 ^0 'V
(ppm)
COD (ppm) 22.24 31.14 22.24 35.58 26.69
NH,.Nitro- .4 .3 -2 .2 -3
.2
.2
.2
.2
57-82
.2
"* gen
S (ppm)
Fe (ppm)
0
<5
0
<5
0
< 5
0
< 5
0
< 5
0
<5
0
<5
0
<5
0
< 5
0
<5
68
-------
Soil and water samples from both tanks were cultured in nutrient
agar after sampling period 5 and all showed profuse colony growth.
Colonies were largely bacilli of at least two different types
as shown by Gram stain.
Indications from these relatively sketchy experiments are that
the use of the porous pavement does not have an apparent in-
hibitory effect on aerobic bacterial processes, and may in fact
promote BOD reduction. More extensive testing, including studies
of out-door demonstration pavements, are needed to remove all
doubts on this point.
69
-------
-------
SECTION XIII
SPECIAL PAVEMENT TYPES
In order for a material to be considered a good candidate for sur-
facing recreational areas at least 4 criteria must be met. First,
the material must be non-abrasive and resilient enough to cushion
falls. Secondly, it must be tough enough to withstand foot traffic
and vandals. Next it should be porous enough so that rainfall drains
through rapidly. Finally, it should be long lived and economical.
Approaches tried by the City of Philadelphia include:
(a) Tanbark matting, which is loose and tends to
spread away from the play area. It is also
becoming expensive and difficult to obtain.
(b) Foam mats, such as Tartan and Safety-Turf poly-
urethane mats, which are laid over black-top.
They are soft and resilient, hold rainwater like
sponges, and are very slippery when wet. Cost
$25-30/yd2.
(c) MonsantoTs Astroturf over black-top like the foam
mats, is resilient, water retentive, $30/yd29 safe
when dry and slippery when wet. Franklin Field
has an 18 inch crown, which still is not enough
to drain the material properly. In the Veteran's
Stadium, a vacuum cleaner is to be used to suck
up water.
A Monsanto M-10 Astroturf was found to have,an impervious closed
cell foam backing. When many holes (l/8"dia., 1/4" apart) were
drilled through this, the piece was amply permeable to bulk water,
but would not drain dry.
In view of the shortcomings of these materials, we took another
approach:
Resilient porous surfacings were made by chopping rubber tubing
into pieces 1/2 to 5/8 inches long, coating them with an adhesive
and pouring them at random into a mold. Mats 1" to 1-1/2" thick
were thus prepared, as described in Table 2. Voids constituted
46 to 64% by volume of the mat, and the equivalent of one inch
of rainfall drained through them in considerably less than 15
minutes.
The mats were all bonded and resilient, and were judged to
possess good mechanical properties for a playground surface.
If damaged by vandals they could be repaired by cementing more
tube pieces into the breach.
71
-------
The samples would weigh 18 to 25 Ibs./yd , and based on the costs2
for bulk rubbers probably would cost a minimum of about $7.00/yd. ,
with $10-12 more likely. Based on commercial extruded tubing
costs, this mat product would cost at least twice these figures.
Adhesives used successfully were Armstrong D-200 cement for gum
rubber, Goodrich A862B Cement for Neoprene, and tetrahydrofuran
for polyurethane. Time did not permit a survey and evaluation of
all the products being offered commercially for playgrounds. A
study of these might disclose materials that could be modified
to make them porous, while still preserving their necessary dur-
ability and resiliency.
Colored Pavements
The economic study (Sec. XVII of this report) indicates a broad
demand for color-infused asphalt pavement, for which a premium
of $0.30 to 3.00 per square yard could be expected. The colors,
preferably light earth tones, are desired to provide a more rest-
ful setting in residential areas, and to eliminate -unpleasant
heat absorption. Brighter colors could be used for color-coding
of deceleration lanes and traffic routings.
Three design approaches to colored pavements were considered:
(a) Use a light colored binder in place of asphalt in
asphalt concrete. Such binders were marketed a few
years ago by Humble Oil and Refining Company, Neville
Chemical Company, and Carlisle Chemical Company.
Three difficulties were encountered:
1. Dirt adhered to the binders, obscuring their
colors. Removal of the dirt was impractical.
2. As a properly-formulated asphalt concrete wears,
aggregate surfaces are exposed and the binder is
confined to the spaces between the rocks. Thus
the dominant color becomes that of the aggregate
itself.
3. Because the colored binders were relatively ex-
pensive, it was desired to use them in thin layers.
Construction of these layers was difficult, and
under continued traffic they tended1to wear away,
exposing the base pavement.
(b) Colored aggregate was used, with asphalt binders, in
normal asphalt concrete formulations. This has been
very satisfactory in localities where a suitable natural
colored aggregate is available (Route 940 near the North
East extension of the Pennsylvania Turnpike, for instance)
72
-------
(c) Colored granules normally used in prepared roofing can be
spread over a resin adhesive layer on the pavement. The
resin can be applied as a hot-melt, cut back with solvent
as a varnish, or emulsified with water. Two such coats
provide excellent hiding of the base pavement.
The color of roofing granules resides in a ceramic layer on their
surfaces. This may tend to chip off under continued heavy traffic,
necessitating a re-application to restore the color.
We obtained a proprietary resin mixture cut back with heptane from
Velsicol Chemical Company, and roofing granules from GAF Corporation,
Hagerstown, Maryland. Two coats of varnish on which granules were ,
spread, were applied to a sample porous pavement of the California
design. They cut its porosity in two, but the resulting product re-
mained sufficiently permeable for any intended installation in the
U.S.
Installed cost is estimated to be 35 to 60 cents per square yard.
Color coatings of various sorts sometimes are used on playgrounds and
tennis courts; these generally are paint-like materials and may con-
tain sand to enhance theirfriction characteristics. Before these
should be used on porous pavements, it should be demonstrated that
they will not cover over or block the pores.
73
-------
-------
SECTION XIV
CONVENTIONAL PAVEMENT COSTS
Porous pavement has been considered for use in constructing the fol-
lowing types of pavements: parking lots., residential and business
streets, county roads and highways, playgrounds, sidewalks, public
squares, and airport runways. Two design levels are considered for
both residential and business streets; county roads and highways are
divided into four classifications based on average traffic volume.
In this Section, standard pavement design, storm drainage provisions
and the installed cost of pavement and storm sewers are discussed for
each of the above conventional pavement categories. Summaries of the
cost and design of these conventional pavements are given in Tables
10 through 12.
In Section XV, "Maintenance and Resurfacing Costs," the average annual
cost of pavement and drainage maintenance and of pavement resurfacing
is assessed to determine the relative maintenance cost range to which
porous pavement must adhere in order to be competitive with conventional
surfaces. In Section XVI, "Porous Pavement Benefits," the economic
1 value of benefits to be derived from porous surfaces is evaluated.
Benefits assessed include relief from combined-sewer overflow pollution,
augmentation of local water supplies, relief from flash flooding and
aesthetic enhancement. In Section XVII, "Market Demand for Porous Pave-'
merit," the gross annual demand for conventional surfaces is estimated,
the cost of equivalent porous design .is compared with the cost of con-
ventional pavements, and the potential demand for porous pavement is
established on the basis of the comparative cost and the benefits to
be realized from porous pavement.
A. PARKING LOTS
Data on parking lots were obtained from a number of private builders
and developers for a standard, large-scale parking lot of approximately
50,000 square yards.
Pavement Design
Because parking lot maintenance and resurfacing are generally the re-
sponsibility of the ultimate owner rather than the developer, parking
lots are typically characterized by a very low level of pavement design.
Standard parking lots are constructed with a 4 to 6-inch thick crushed-
stone base and a 1-1/2 to 2-1/2-inch thick asphaltic surface. In urbanized
areas cities and Standard Metropolitan Statistical Areas where speci-
fications are stricter, a 6-inch base and a 2 to 2-1/2 inch surface are
normal. In outlying areas where specifications are less stringent,
5-1/2 inches in total thickness is typical.
75
-------
Base costs have been estimated from primary and secondary sources
at from $1.30 to $1.60 per square yard and surface costs at from
$1.30 to $1.90 per square yard, yielding an average total pavement
cost of $3.05 per square yard. Although smaller parking lots have
the same basic design, the unit installed cost is inflated because
of the small surface area. Costs such as excavation and grading
have not been considered in this and subsequent analyses since they
are invariant to the pavement type.
Storm Sewers
Discussions with developers have established that storm-sewer facil-
ities are almost invariably required by local building codes for
large-scale parking lots in both urban and outlying areas. In
standard practice, one storm inlet is provided for every 2000 square
yards of surface area. When unusual terrain and/or runoff prevails,
more inlets may be required. The linear feet of storm drainpipe
varies with the locale and depends on whether runoff is directed to
an existing nearby trunk sewer system or outflowed directly to a re-
ceiving watercourse. Estimates obtained from the City of Philadelphia,
the town of Columbia, Maryland, and several developers indicate that
the cost of storm sewers ranges from $1.05 to $2.06 per square yard
of surface area. The low figure is typical of an inner city parking
lot where little piping is required to link storm inlets with an
existing street sewer system; the high figure represents an outlying
area where more extensive provisions are required. When a high water
table or other local condition necessitates the removal of subsurface
water from the paved area, underdrain pipes are sometimes installed
at an extra cost of approximately 15£ per square yard.
B. RESIDENTIAL STREETS
Since the basic construction of residential streets varies consider-
ably, residential roads were classified by design level low design
and high design and are discussed separately below.
Pavement Design
Low-design residential streets are typically located in suburban and
outlying developments where county specifications are less stringent
than those in cities. Typical construction consists of a 5 to 6-inch
fo crushed stone with a 1-1/2 to 2-inch asphalt surface. The low end of
this range 6-1/2 inch total thickness is generally observed in such
outlying developments as Columbia, Maryland; the higher level of the
design range 8-inch total thickness is typical of suburban counties
within metropolitan areas. Total cost for these low-design residential
streets ranges from $3.00 to $3.60 per square yard. High-design re-
sidential streets are generally only required in metropolitan areas.
In these areas high standards are enforced so that roads will withstand
the heavy traffic volume generally associated with dense urban residential
development without necessitating heavy maintenance expenditures. These
roads typically consist of a 4-inch crushed-stone subbase, a 6 to 8-inch
76
-------
asphaltic concrete base, and a li-inch asphalt surface. Based on bid
figures provided by the City of Philadelphia, the cost of such streets
varies, from $4.75 to $6.10 per square yard.
Storm'Sewers
Road storm sewers are required for almost all developments and
urbanized areas. The design standards for such systems,1and thus their
cost, .are, however, subject to considerable variation. High-design
systems are generally installed in large-scale, new town developments
and city residential areas. In high-design systems, inlets are spaced
about every 200 feet along the road length with parallel storm drain-
pipes throughout. Double inlets ,are generally provided where resi-
dential streets join townwide roads. Data supplied by the City of
Philadelphia, the New Communities division of HUD, the town of Columbia,
Maryland, and the urban drainage district of Denver for a number of
large .residential develop'ments indicate that high-design residential
road sewer systems cost from $6.82 to $10.04 per -front foot of building
lot, or from $4.09 to $6.02 per square yard of surface area for a stand-
ard 30-foot wide residential street. In low-design systems, inlets are
spaced further apart and ! the length of storm drainpipe is reduced. In
very-low design systems, only one inlet ijs provided every few blo.cks
and stormwater runoff is .allowed to cross1 street intersections. !Since
inlet locations are designed to minimize the distance to existing trunk
sewers or receiving water-courses, sewer pipes are not necessitated
throughout the entire street length. The cost ofi such systems typically
ranges, from $1.80 to $3.00 per square yard of road surface.
Cost of Curbs and Gutters
Combined curbs and gutters are generally installed in new residential
developments at 4 cost.of from $2.50 to $4.00 per linear foot. Eor a
30-foot-widejresidential road, curbs and gutters cost an additional
$1.50 to $2.40 per square yard of surface area, which could be elim-
inated if porous pavements were used.
C. BUSINESS'STREETS , ' : . ' . -
Whereas suburban business streets are typically townwide roads designed
to handle only a relatively moderate volume of traffic, city business
streets must,be designed :to withstand greater traffic volumes. Since
the design of business streets varies considerably with the expected
traffic volume, city and suburban business streets are discussed sep-
arately below. '
Pavement Design
Suburban busi.ness streets are typically constructed with, a 6 to 7-inch
crushed-stone base and a 2 to 2i-inch asphalt surface at a total cost
of between $3.60 and $4.40 per square yard. To support considerable
volumes of traffic, city 'business streets are generally constructed
77
-------
with a 6 to 10-inch concrete base, the thickness varying with the
anticipated traffic volume, and a 2i-inch asphalt surface. Bid
data supplied by the City of Philadelphia indicate a pavement cost
of from $5.22 to $10.20 per square yard for these very-high-design
roads.
Storm Sewers
Developers contacted indicated that the unit cost of storm-sewer
systems for suburban business streets average from 20 to 25 percent
higher than for high-design residential street systems. For a typical
40-foot-wide suburban business street, the average cost of storm drain-
age facilities would thus range from $3.30 to $5.50 per square yard of
road surface. That the cost per square yard is less than for high-
design residential systems is consistent with the fact that storm-sewer
system costs do not increase proportionally with the increase in volume
of surface runoff. In this instance, unlike city business streets,
the increase in runoff is primarily due to the increase in road width
rather than to increases in the extent of adjacent imprevious areas.
Data on the cost of road storm-sewer systems for city business streets
vrere extremely difficult to obtain.. Only one engineering firm was
willing to provide an estimate of approximately $9.00 to $12.00 per
square yard. This high cost is consistent with large increases in run-
off from surrounding impervious areas and the normally closer spacing
of storm inlets in city business areas.
I
D. COUNTY ROADS AND HIGHWAYS
The classification and design criteria for county roads and highways
has been based largely on Pennsylvania Department of Highways specifi-
cations. Information from secondary sources and other state highway
departments indicates that these specifications represent standard
practice except in the case of low-level design county roads with
traffic volumes of less than 800 cars per day. In most states, roads
with an average daily traffic (ADT) of between 100 and 800 vehicles
are designed to lower specifications than in Pennsylvania, and rural
roads of less than 100 ADT are typically only treated gravel. Accord-
ingly, the Pennsylvania classification system and design specifications
have been adjusted to reflect lower standards in roads with 100 to 800
ADT; roads of less than 100 ADT, being generally unpaved, have been
excluded from consideration.
Based on the above modifications, four county road classes have been
been identified.
Road Class
Low-volume county road
Moderate-volume county road
Two-lane highway
Four-lane highway
Average Daily Traffic
100 to 800
800 to 2000
2000 to 10,000
above 10,000
78
-------
Pavement Design
Design standards for county roads and highways by road class are
indicated below:
Low-Volume County Road ^
Subbase: 6" (crushed stone or gravel)
Base: 4i" (bituminous concrete)
Surface: li" to 2i" (bituminous wearing and binder course)
Moderate-Volume County Road
Subbase: 8" to 12" (crushed stone or gravel)
Base: 5" to 6" (bituminous concrete, bituminous aggregate
or cement aggregate)
Surface: 2i" (bituminous wearing and binder course.)
Two-Lane Highway
Subbase: 8" to 15" (crushed stone or gravel)
Base: 5" to 6" (bituminous concrete, bituminous aggregate
or cement aggregate)
Surface: 2i" to 4" (bituminous wearing and binder course)
Four-Lane Highway
Asphalt
Subbase:
Base:
Surface:
Concrete
Base:
Surface:
8" to 16" (crushed stone or gravel)
7" to 10" (bituminous concrete, bituminous aggre-
gate or cement aggregate
3%" to 5" (bituminous wearing and binder course)
8" to 12" (crushed stone, gravel or slag aggregate)
8" to 10" ( plain cement concrete or reinforced
cement concrete)
Four-lane concrete highways are generally designed with surface and
base only. However, in some states, particularly, a 4-inch cement-
treated base and a thinner subbase are used. Cost for such construc-
tion is only slightly higher than for the above design.
The following total cost ranges have been calculated from a sample
of bid sheets provided by the Pennsylvania Department of Highways.
Since bids by contractors on the same work will vary by as much as
100 percent depending on the relative state of the industry, extremely
high and low bids have been eliminated in arriving at these cost ranges.
(Costs for individual road elements are detailed in Table 10.)
79
-------
Type of Road
Low-volume county road
Moderate-volume county road
TWb-lane highway
Four-lane highway
Asphalt
Concrete
Storm Sewers
Pavement Cost ($/y )
5.30 - 7.00
6.35 - 8.75
6.50 - 10.60
8.60 - 12.50
10.00 - 15.00
Three types of drainage systems are provided in county roads-and
highways: underdrain pipe to remove groundwater from the road
base and prevent freeze-thaw damage, storm drainpipe to remove
surface runoff, and culvert pipe to channel water under the roads
where an existing stream or watercourse is intersected. Because
culverttpipes would normally still have to be installed if porous
pavement were used, the cost of culvert pipe installation has not
been considered in this analysis.
The extent of surface and underdrainage facilities installed depends
on the design level of the road itself. The higher the orginal
design of the road (i.e., highway versus low-volume county road),
the higher the level of drainage facilities provided because of the
higher replacement cost of the road and because most higher design
roads must meet federal specifications. (Most highways are funded
through the Federal-Aid'Primary or Interstate systems.) The frequency
of inlet spacing between low-level county roads and highways typically
differs by a factor of 10. Additionally, unlike small developments
x*hich tie into existing trunk sewer systems, major roads must provide
their own capacity to the point of discharge. Where the distance to
the discharge point is great, end-of-run pipes can reach over 72 inches
in diameter, resulting in very heavy excavation and backfill costs.
In large-scale residential and commercial developments, layouts are
planned to avoid such excessive storm drainpipe costs. Further, within
equivalent design classes, drainage provisions also vary as a result of
difference in anticipated extent of runoff, road terrain, height of
water table and anticipated amount of precipitation.
To arrive at reasonably representative cost figures, drainage costs
including excavation and backfill for each county road and highway
class have been calculated from a sample of construction bid figures
provided by the Pennsylvania Department of Highways. These figures
are representative for most of the country except for areas of ex-
tremely low or extremely high precipitation. In these areas, some
adjustment would be necessary to reflect significant differences in
size of storm drainpipes required to handle the anticipated volume of
runoff. However, basic system design, which is the major determinant
of drainage cost, is relatively uniform throughout the country. The
80
-------
The relative insensitivity of systems design to annual precipitation
is indicated, for example, by the fact that the residential storm
drainage system design recommended for Columbia, Maryland, and
Denver, Colorado, are essentially equivalent although the size of
pipe required to handle anticipated runoff differs.
From the above Pennsylvania data, the following total drainage cost
statistics (surface and underdrainage) per square yard of road sur-
face have been calculated for county roads and highways.
Type of Road
Low-volume county road
Moderate-volume county road
Two-lane highway
Four-lane highway
Total Cost
Range ($/y ) Weighted Average ($/y2)
0.68
2.15
2.04
7.76
4.27
6.43
9.85
18.90
1.27
3.43
5.95
13.33
Total average costs per square yard for pavement and drainage pro-
visions on county roads and highways are indicated below:
Type of Road
Low-volume county road
Moderate-volume county road
Two-lane highway
Four-lane highway
Asphalt
Concrete
Sidewalks
Z 9
Range ($/y ) Weighted Average ($/y )
5.98
8.50
8.54
16.36
17.76
11.27
15.18
20.45
31.40
33.90
7.42
10.98
14.50
23.88
25.83
Basic sidewalk construction consists of a 2 to 4-inch gravel base
and a 4-inch concrete surface. Occasionally, a 6-inch surface is
used. Because of the small surface area and hand construction in-
volved, costs per square yard of surface area are quite high, typi-
cally ranging from $5.70 to $8.05 per square yard. Drainage systems
are not normally necessary.
E. PUBLIC SQUARES
Pavement Design
Public squares are normally constructed with a 4-inch gravel base
and a 4 to 6-inch concrete surface, 6 inches generally being employed
only where some vehicular traffic or other special factors are antic-
ipated. Costs range from $5.05 to $6.55 per square yard.
81
-------
Costs are relatively cheaper per unit than for sidewalks because of
the larger average surface area and more mechanized construction,
e.g., belt dragging rather than-hand finishing.
Storm Sewers
The expected volume of storm runoff for public squares is essen-
tially equal to that for parking lots. Storm drainage system
design is thus equivalent to that for parking lots, with inlets
normally spaced every 2000 square yards. Total system cost is
somewhat lower, ranging from $0.90 to $1.20 per square yard of sur-
face area, because proximity to existing street sewers reduces the
length of storm drainpipe required.
F. PLAYGROUNDS
Information supplied by state and local recreation authorities in-
dicates that playground design is relatively standardized. Most
playgrounds are constructed with a 4-inch stone base and either a 2-
or 24-inch asphalt surface. Pavement costs for standard playgrounds
range from $3.00 to $4.00 per square yard. Because of the smaller
surface area of playgrounds, unit costs for equivalent designs are
higher for playgrounds than for parking lots.
Storm Sewers
Costs of storm drainage facilities were estimated to range from $0.90
to $1.20 per square yard. Drainage provisions in urban areas are
generally equivalent to those for public squares; standards in out-
lying areas are somewhat lower, but greater pipe footage is often re-
quired to connect with street storm-sewer lines.
G. AIRPORT RUNWAYS
Pavement Design
The design of major airport runways has been established for asphalt
and concrete and is based on information from the Philadelphia
International Airport, the Portland Cement Association and the Federal
Aviation Administration's Bureau of Airports. Asphalt runways are
typically constructed with a 3 to 12-inch crushed-stone subbase, a
7 to 12-inch bituminous penetration base, and a 4 to 6-inch asphalt
surface. Variation in the thickness of construction materials depends
both on the design level of the runway and the segment of the runway
being considered, that is, the critical or non-critical portion of the
runway. Major asphalt runway costs range from $8.00 to $12.90 per
square yard of surface area.
Typical concrete runways for major airports consist of a 3 to 12-inch
crushed-stone subbase, a 4 to 6-inch soil-cement base, and a 9 to
13-inch reinforced concrete surface. For concrete construction, sub-
bases are generally limited to 3 inches unless greater thicknesses
82
-------
are necessary to prevent freeze-thaw damage. Costs for these runways
are estimated at from $11.35 to $18.80 per square yard of surface
area.
Storm Sewers
Runway storm drainage systems are constructed to avoid placing inlets
in the runway surface itself. Accordingly, runways are generally
sloped sufficiently to allow runoff to drain into inlets placed in
ground areas bordering the runway. Cost statistics supplied by
Philadelphia International Airport indicate that the total cost for
such a system (subdrains, drainage structures and drainpipes) is
$1,330,000, or $7.60 per square yard for a standard intercontinental
runway 10,500 feet long and 150 feet wide.
H. SUMMARY OF CONVENTIONAL PAVEMENT COSTS
Pavement and drainage costs discussed in the text are summarized,in
Tables 10 through 12. Tables 10 and 11 give the range of cost estimates
established for pavement and storm-sewer installation respectively.
Average total costs for both pavement installation and drainage pro-
visions are summarized in Table 12.
83
-------
CO
c
o
i i
s
c?
i
o
(
0 g
UJ
01 5;
1 «,£
(O D*
r _j
CM
^
CO-
ft
in in
fl) QJ
c j:
V C
u c
1
CM
*
in..
in
4-1
C
Q)
0)
(0
Q_
M-
O
4>
p
O
in
CO
i
o
vO
CM
CM
r
t^S*
o
cn
i
0
CO
vC
1
-3-
o
vO
r
|
o
CO
I
1
1
4->
O
I
cn
c
ID
0-
o o
vO
CO vQ
1 1
o tn
o r-.
co -3-'
CM
1
^-. r_
o in
cn CM
i i
in o
in o
VO CO
1 1
tn vo
o tn
P- -3-
CO
1 1
in tn
-*r I"***-
CM
-3"
O
,
1
O
O
a)
(U
4-J
01 C
C 01
< cn
(D in
~ in a)
4-1 (U "O
c -a i
in i u:
a)
o
O CM
-3"
O
-3"
1 1
o r^
vO CM
co in
CM
i
CM CM
O O
-3- O
CM CM
1 1
O O
on tn
;
o
i i
vO VO
O O
O CM
CM CO
1 1
o r-
CO
4-1
c
D.
o
1_ QJ
4-J
co c
3 -£
'u} ^ 0
CO
o in
o r~
r~- co
i i
o m
CO CO
LA VO
A"
CM
1
CM
O LA
O CM
CM CM
1 1
O LA
'.C
^- '
LA
0 0
O LA
CO CO
1 1
O O
i^- vO
CM CM
CM .
1
vO CO
O O
0 0
CM CO
1 1
0 0
CO O
^ CM
i
,
O
0) >
-o E i
ID 3 (U
O t->
c£ O CD
> i-
>- 1 (U
4-J 3 "O
COO
3 _i z:
o
0
O O 0
VO LA O
O CM LA
r 1 r
1 1 1
O O O
LA VD O
vO CO O
-3 LA O
1 1
1
CM CO CO
o
o o o
co co
CM
CM CO
1 1 1
o o o
1 Q. C
to i i- m o
2 O 3 < o
.c 5 o
cn I U-
=c
LA
O
co'
1
O
r»
tn
-3"
O
O
r-^
I
LA
-3"
-3"
1
CM
LA
O
r
\
LA
I .
0
li
to
I
-o
CO
LA
LA
VO
|
LA
O
LA
vD
1
-3"
O
LA
LA
1
O
O
3-
"*
in
o
-
fD
Q.
o o
cn °°
CM °°
"~~ '
i i
o u"*
O fo
CO '~~
VO
1
.3- on
t
o
o o
°°. -3-
CO
1 1
rC §
CM cn
.,
CM
i I VO
1 1
r-~ ^r
2 S
vo'
1 1
0 0
CM
-3"
CM £J
1 1
CO ^
1
o o
o o
ro co
i i
in LA
^
CD
S
c
3
^ 4-1 4-1
4J i-l a)
° 1. §
£ 3 5
"*
X
OJ
T3
0)
(D
O
T3
C
84
-------
O)
1
_J
tu
o
ec£
ca
Q
"z*'
ea-
_T
0
I I
t
£
u^*
c3^
o
X
UJ
«*
^*"
o
t-H
1
3
5:
to
i i
CD
2:
I l
O
ID
ji-J
C_!>
s:
I-H
f\
to
l
o
o
to
UJ
1 1
1
1 1
_J
O
^^
U-
UJ
CD
-
O
H- '
>
S- "V.
-o -
!U. *- ^
0)
T3
C
^
0)
o ^-^
(D CM
M- >
&_ ^^
3 -tO-
CO
4-1
C
(U
E
0>
>
03
Q.
It-
O
fl)
Q.
>^
i-
^
CM
CM
1
LA
o
.
LA
.
.
o
i
o
VQ
o
CM
I
LA
o
i
4-J
O
en
c
,
*s
U
03
Q-
CM CM
O O
VO vO
1 1
O 0
00 VO
. PA
'
1 1
I |
CM CM
O 0
vO vO
1 1
O O
CO VO
*~~ PA
4-J
0)
0)
1_
4-J
to rr
c en
en
03 in
in (U
4-J (D "O
c -a i
0) (U '
u -o
4-1
to c
1C
in -Q
in u
<1J 3 ^
C -Q 4-J
3
in in o
3
CO
-3> *3c
r^ PA
CM -3-
_cj- vO
1 I
co LA
vO
O CM
o~^ r~>
. .
CM O
r i
CD vO
-* CM
o o
O CM
OO .
o -=r
i i
0 VO
CM 0-<
O O
(1)
E
3
^
O
(1) >
"o E i
03 3 (D
O 4-1
Oi O 03
> U
>- 1 0)
4J 3 -a
coo
3 _1 S
O
-X
?c O
LA CTv
OO
OO
CT\
1 1
-a- vo
CD r
CM r-^
PAvD
« CM
PA
1 l
LA r^-.
VO OO
.
o o
_^-
vO PA
r *
PA
vO
1 1
-cr -^r
PA
o LA
0)
* *~~ 1
03 1 >-
303
-c 3 O
cnf LU
«
n:
a)
c
o
^~
0)
c
o
z.
0)
c
o
\s
^~
03
3
(l)
to
o
CM
.
1
O
O1
O
1
1
o
CM
r^
1
CD
CP\
O
Q)
1_
03
^3
CT
CO
U
.
-Q
^
°-
O
CM
,
1
O
CTl
O
I
i
O
CM
,
I
O
en
»
o
"D
C
0
en
>^
03
^_
Q.
0
VO
r*%
I
I
1
1
^>^
03
3
c
3
rf
4-1
L-
O
Q.
.
<
^
4)
03
03
Q-
0)
in
"a
4-1 -a
m c
O 3
o
-a
U) C
a) 03
-a
3 (I)
o
U 03
C 4
^~
3
in
fi3
O O
5^
85
-------
CM
O)
o
14
O
r\*
f
CO
^"
o
O
1
2:
UJ
«
UJ
Q_
in ^-»
O CM
0 >-
ro
4-J
o
H-
E
O
to
1 i
c
E CM
Q) >
> ^s.
ro -co
a. '
4-J
c
ox
0)
ro
o
0)
D-
CA
^t
CO
vO
«
1
LA
O
CA
4J
O
en
c
ro
O-
O CA
CM CM
r^- CD
o o
en oo
CA -^3"
O CA
CA LA
j i
0)
0)
L.
4-1 C
co c en
en
in
ro m TD 1
C 1 -C
a) 3 en
-oo
in
0)
c£.
O O
-^- CM
co oo'
o o
-^- LA
-a- o
0 0
o r^-
-3-" r~~-
c
0)
E
Q.
0
(D
0) 0)
to ~a
4-J C
co ro
-Q
if> i-
m 3 >-
0) JD 4->
C 3
to cj
in
CO
CM OO
-3- CTl
r*^ o
CM -*
*
< CA
"
LA LA
LA
\0 t~^~
0)
E
13
County Road
Low-vol ume
Moderate-vo
O OO CA
LA OO OO
-3- CA LA
CM CM
LA CA CA
CA CA CA
LA CA CA
r~~" *"""
LA LA O
LA LA LA
00 O CM
*""" '
Highway
Two- lane
Four-lane
Asphalt
Concrete
CO
CO
vJD
a)
C
O
z.
CO
co
MD
Sidewalk
LA
DO
^
LA
O
DC
LA
-------
SECTION XV
MAINTENANCE AND RESURFACING COSTS
Two aspects of recurring costs must be considered in assessing the
economic feasibility of porous pavement: (1) the cost and required
frequency of resurfacing and (2) the cost of routine annual main-
tenance. In characterizing costs between these two categories we
have, with minor exception, followed the definition of the American
Association of State Highway Officials which classifies surface treat-
ment of less than threequarters of an inch as maintenance, and of
more than three-quarters of an inch as resurfacing.
A. COST AND FREQUENCY OF RESURFACING
Road Service Life
Data on resurfacing practices and the service life of original road
surfacesthe time until the first resurfacing is requiredhave been
obtained from a number of sources, 9-12 including city, state and pri-
vate sources, and are summarized in Table. 13. Because of their ex-
tremely low-level original design, parking lots generally require
resurfacing after six years. Low-level design, flexible roads (low-
design residential, business and county roads) have a service life
of about 15 years, whereas high-level design, flexible surfaces last
about 18 years and concrete roads about 26 years. Sidewalks and
public squares require resurfacing after 20 years only if the pavement
is eroded by snow and ice removal chemicals.
Resurfacing Practices
An asphalt overlay is almost universally used for resurfacing asphalt
and concrete pavements. The thickness of the resurfacing layer nor-
mally varies between 1 and 3 inches (see Table 13), except for concrete
runways where an average 4-inch overlay is used. As a rule, the thick-
ness of the overlay varies linearly with the quality of the original
road design. Thus a high-level design concrete road is generally given
a 3-inch resurfacing, whereas low-level design residential streets may
receive as little as a 1-inch overlay. For both sidewalks and public
squares, resurfacing involves the complete removal and replacement of
the original surface.
Resurfacing Cost
Total resurfacing costs, derived from secondary studies, are summarized
in Table 1.3. Annual resurfacing costs have been derived on the basis
of the length of time to first resurfacing and the interval between
successive resurfacings evaluated over a 40-year period.
12
Based on information from the California Highway study, the time
interval between resurfacings has been assumed to be 13 years for
roadways. Resurfacing intervals for parking lots and playgrounds have
87
-------
to
o
o
o
£
UJ
U.
LU
O
LU
to
^;
in
OCvT
_ -^
ro -
O -»
o
^
in
c
o
-C'
1- w
0)
01 jr
c o
c
o
ta*'
L.
3
in
»
(1)
4-1
C
H)
ro
o_
0
u
a.
1-
VO
CM
O
t
CO
CM
O
LTi
-3-
^
1
LA
CM
CM
1
^~
v£>
0
cn
c
OO CM
O
0 O
0 0
1
^
CM
LA
LA f~
4-1
O
(U
L_
4J
Residential S
Low-design
High-design
CO CM
0
o o
0 0
r 1
S£J
CM
LA
LA r--
4-1
C
Q.
0
4-* 0)
a) >
Business Stre
Suburban De
City
o
1
cnco
o o
O 0
o
i
_
i
LA LA
CM CM
A*1
CM
I
\p* ?\w
r 1
LA CO
0)
e
3
County Road
Low-vol ume
Moderate-vo
o o
I 1
CO OO
o o
o o o
o
_
1
LA O LA
CM r- - o
CM
A"
CM
1
f^1 T^?*
CM ro
OO CO VXD
CM
Highway
Two- lane
'Four- lane
Asphal t
Concrete
5-
0
i
r^,
O
0
CM
OO
1
LA
VO
-3-
O
- ro
>- i-
-Q O
01 >.
i
in ro
-o E
O l-
o
1 U
ID .-
a. c
3
in
<
in
>-
^
ro >-
ro 5
i- in
ro
O m
>-
1 -C
O 4-"
-3-
C
^ *^
3
O
M-
4J
ro
Dl
C
>-
._
U
ro
o
14-
0
in
4_j
in
O
o
in
X -1"
ro
a.
n>
u
4)
x:
2
>-
c
O
^,
u
ro
m
in
a)
o
o
C
a)
L-
L.
0
14-
CM
>-
~^.
O
CS
1
14-
o
4-)
(/I
0
O
l/l
?u
D
D
O
C
_^l
*o .
88
-------
been determined to be 6 years and 12.5 years, respectively. Side-
walks and public squares need replacement about every 20 years if
eroded by snow and ice removal chemicals or otherwise damaged. By
FAA standards, airport runways are designed for a 20-year life. In-
creasing traffic volumes and heavier jets have precluded this life
expectancy from being realized. According to Philadelphia Airport
engineers, the initial service life of airport runways is estimated
at 10 years for asphalt construction and 15 years for concrete
construction, with subsequent resurfacings required at 10-year intervals
B. ANNUAL MAINTENANCE COST
Annual maintenance provisions considered in this report can be classi-
fied according to three major types of activities: (1) repairs to
surface and base, (2) surface overlay treatments less than three-
quarters of an inch thick and (3) repair and maintenance of drainage
structures. Other maintenance activities such as control of roadside
vegetation, snow and ice removal and repair of rest areas, although
constituting a sizeable fraction of total maintenance costs, have not
been considered because they are not relevant to the type of pavement '
surface. The three maintenance categories considered are described
below, and costs for each type of pavement are given in the maintenance
cost section.
Repairs to Surface and Base
Surface and base repair is occasioned primarily by four types of re-r
current road defects: (1) potholes and chuckholes, (2) cracks and
other minor varieties of surface disintegration, (3) edge curling and
other edge damage and (4) pavement pumping. The first two types of
surface damage occur because of inadequate base construction, poor
drainage, and/or heavy traffic. Edge damage is typically encountered
only in roads less than 24 feet wide where cars frequently travel over
the edge. Pavement pumping occurs when concrete slabs are separated
and base or subgrade materials are extruded through pavement joints.
Surface damage of the first three types is corrected by adding rolled
asphalt with or without additional aggregate, depending on the depth
of the damage. To prevent pavement pumping, concrete joints are
generally sealed and waterproofed periodically. If pumping occurs, the
pavement is packed with a mud slurry under the surface level and, if
necessary, the base and slab are replaced.
Surface Treatment
Pavement surfaces become worn and-smooth before they require complete
replacement-, thus increasing the danger of accidents and also exposing
the underlying base materials. To correct this condition, pavements
are periodically sealed with an asphalt coat, generally less than
three-eighths of an inch thick, between the time of initial construction
and resurfacing as well as between successive resurfacings.
89
-------
Maintenance of Drainage Facilities
Maintenance of drainage facilities comprises the following activities:
(1) repair of drainage structures Such as inlets, manholes^ and drain-
pipes (2) unclogging of storm inlets and (3) repair of drainage channels.
Maintenance Cost
In determining total maintenance cost, separate costs have been estimated
for each of the three major maintenance activities above* The paucity .
of usable maintenance data has prohibited us from determining the quan-
tity and unit cost of each maintenance activity as preferred. Rather,
typical costs have been derived on the basis of reported maintenance
cost experience for the three categories above* Because no one sotitce
provided adequate data to estimate costs for all pavement categories
arid maintenance types> data were obtained from Several sources (includ-
ing state and local officials arid private contractors) which, in our
opinion, reflect typical levels of maintenance activities.1^1 The
costs thus derived on the basis of direct labor^ equipment and materials
costs are summarized in Table 14, and the derivation of the cost figures
for each pavement type is detailed in the paragraphs below*
Parking Lots
Information from local paving contractors indicates that, annual
surface and base maintenance -, primarily pothole repair> requites the
use of 75 to 100 tons of asphalt materials at an installed cost of
$2250 to $3000 per year for a standard 50,,000 squares-yard parking area.
On a square-yard basis, the animal surface and base maintenance Cost is
4»4
-------
CO
o
o
UJ
C_J
CO
-P
OJ
0
c
03
C
a>
n 4-cM
ui >-
03 O ^~
2: o -o
_
03
O
t
4-1
C
E
-
0) -u-
0 03 ~ '
03 a;
y- co
CO
(U
U
a) c
cn 03
03 CCM
c a) >-
03 C -0
O 03
2:
a)
in
m
CO
UI
- ex?
I) 03 >-
O Q.^-,.
03 a) -o-
L-
co
4-J
c
E
4)
O_
^-
O
a)
Q.
h-
0
0
xQ
1
O
-3"
^f
t_ ,
03
C
O
o
xO
1
O
"*
O
_J
Dl
C
Q-
OO -3-
cxi cxi
' CM
CM O
-"' '
-3-
OA -3-
0 O
OO O
CXI O
"""
41
a)
a)
L-
4-J
CO C
C Dl
Dl
03 l/l
- ui a)
j-> a) -o
C "D 1
(U 1 -C
-0 3 cn
o
Ul 1 31
1- TD
usiness St
Suburban
City
CO
OO
xO LA
O
1
, LA
r-~
OA
CM-
CM O
'
CM
OO OO
O O
OA
CO
-3- CM
O
1
O
'
a)
E
3
O
aj >
ounty Road
Low-vol um
Mo derate -
o
CO xO
cxi r-
CM xD -3-
cn
|{
-3"
O*\ f^.
LA O
LA -3-
CM CM vO
o o cn
0
LA LA
0 0
CO '
CO II
O xD xO
CA CA
0 0
CM r- LA
O Cxi 1
PA -3- CM
C
cn
i/>
5)
-a
j
cn aj
aj j-< *j
a) c _rr ^~ . jz o
>- 1 L. Q. c
1T3 1 L_ i) U^ O
2 O 13 > < 0
-c 5 O '
cn 1 u_
03
-a
£
03
E
C
2:
d3
L.
m
D
cr
CO
o
_a
a.
P;
.
03
0)
>
O
03
-C
Q.
I/I
03
a)
c
o
JZ
c
I
i_
a;
03
Z!
CT
I
.c
91
-------
surface treatment is 1.210 per year. On the basis of the Virginia
study, 14 the cost of drainage maintenance is estimated at 0.340 per
square yard. This figure covers only the cost of cleaning and re-
pairing drainage structures as opposed to maintenance for county
roads and highways where considerable amount of drainage maintenance
expenditure reflects the upkeep of drainage channels.
High-Design Residential Streets
Specific data on surface and base repair cost for high-design re-
sidential streets were not located. In view of the superior design
of these streets, particularly the quality of the base, we have
estimated the annual maintenance cost at 1.00 per square yardsome-
what lower than for low-design residential streets. Drainage main-
tenance cost was estimated from the Ohio survey ' at 410 per square
yard on the basis of cost data for local urban bituminous roads.
Based on an original service life of 17.5 years and a 40-year analysis
period, the cost of seal-coating was estimated to average 1.020 per
square yard per year.
Suburban Business Streets
Surface and base maintenance was estimated at 1.490 to 1.620 per
square yard per year, equivalent to the maintenance cost of a low-
volume, 40-foot-wide county road with an average daily traffic of
from 600 to 800 vehicles. Drainage maintenance cost was estimated
from the Ohio study15 at 0.590 per square yard per year on the,basis
of cost data for auxiliary rural bituminous roads. Cost for one
surface treatment between resurfacings was estimated at 1.210 per
square yard per year.
City Business Streets
Data on the cost of surface and base repairs for city business
streets in Philadelphia was obtained from the Pennsylvania.Department
of Highways for fiscal years 1966 and 1967. Inflated to current price
levels by the Federal Highway Administration's maintenance price index,
the cost was estimated at $648.32 per lane mile or 9.210 per square
yard per year. Data from the City of Philadelphia indicated that sur-
face treatment consisted of a three-quarter-inch to one-inch overlay
(rather than the thin layers generally used) at an average annual cost
of 6.050 per square yard per year. Annual drainage maintenance cost
was estimated from the Ohio study at 0.790 for urban arterial
bituminous streets.
Low-Volume County Roads
Annual surface and base repair cost was estimated from National
Cooperative Highway Research data17 at 1.700 to 4.830 per square
yard for a 22-foot-wide road with ADT from 200 to 800 per day.
92
-------
The higher cost than for residential and suburban business streets
reflects the cost of repairing edge damage, a cost which is gen-
erally only significant for narrow, uncurbed roads. The range of
maintenance costs reflects the tendency of edge repair costs to
escalate more rapidly than other costs with increase in traffic
volume. In the Virginia maintenance study,14 Jorgensen and
Associates estimated annual drainage maintenance costs at 1.820 per
square yard of road surface. This cost reflects drainage-channel
upkeep as well as drainage-structure cleaning and repair. Based on
a 15-year initial life, surface treatment cost was estimated to
average 1.210 per square yard per year.
Moderate-Volume County Roads
Surface and base repair costs, adjusted to current price levels,
have been estimated at 3.020 per square yard on the basis of
historical per lane mile maintenance costs for Pennsylvania high-
design flexible roads. Drainage maintenance cost was estimated at
0.810 per yard using Jorgensen data. Based on an initial service
life of 18 years, surface treatment was estimated to average 1.020
per square yard per year.
Two-Lane Highways
Surface and base repciir costs for Pennsylvania high-design flexi-
ble roads were used to arrive at an estimated annual maintenance
cost of 3.020 per square yard. Drainage maintenance costs based
on the Virginia study data-L4 were estimated at 0.880 per square
yard. Surface treatment, 18-year initial life, was estimated to
average 1.020 annually per square yard.
Four-Lane Highways
Pennsylvania maintenance data did not provide maintenance cost
estimates for very-high-design four-lane highways. Surface and
base repair costs were therefore estimated from the very compre-
hensive California highway maintenance cost study.12 Adjusted to
current price levels, surface and base repair costs from this source
were estimated to average between 4.210 and 2.750 per square yard
for asphalt and concrete construction, respectively. The higher main-
tenance cost for asphalt highways reflects the greater average dura-
bility of concrete construction. Annual drainage maintenance costs
were estimated from Virginia maintenance data-*-4 to range from 0.360
to 1.050 per square yard. This wide range reflects the considerable
maintenance economies realized where median-strip drainage is employed
on divided four-lane highways. Surface treatment was estimated at 1.020
and 0.960 per square yard per year on the basis of 18 and 26-year service'
lives for asphalt and concrete highways, respectively.
93
-------
Sidewalks
Surface repairs, other than replacement, and surface treatment
are generally not made* Since drainage is typically into streets
rather than through sidewalk inlets, no direct drainage maintenance
cost is incurred.
Public Squares
Surface repair or surface treatment is not normally undertaken.
Drainage maintenance cost was judged insignificant for the same
reasons given under parking lots.
Playgrounds
Data from the Philadelphia and Pennsylvania Department of Recreation
indicate that surface and base repair activity is very limited. Re-
pair is generally restticted to one thorough patching job between
resurfacirigs. On the basis of parking lot repair costs, such a
patching job would cost 0.37£ per square yard oil an average annual
basis. Surfaces are generally not treated between resurfacings.
Since it is rarely necessary to clear a clogged tiilet, drainage main-
tenance costs are insignificant.
Airport Runways
No specific cost data were available on surface and base repairs fbir
airport runways. Contact with Philadelphia International Airpo-rt
personnel indicated that maintenance activity was similar to highway
repairs, consisting primarily of repairing potholes oft asphalt run-
ways and sealing and waterproofing joints on concrete runways. Con-
sequently, maintenance cost for runways has been estimated to be
equivalent to that for high-design highways: 4.21C and 2.75
-------
OO
oo
CD
O
CJ3
i i
C_5
C£
--'- -"J
OO
LLJ
C£
LD
r Q
l~> LJJ
03 <-3
LJJ
t
1 1
2:
CD
M-
O
(D
Q.
H-
O
CM
CO
1
-cr
CM
o
C71
C
s_
CT3
Q_
CO -3-
O -3"
i i^
4-J
tt)
CD
!__
4-J
CO C
c cn
CD
CD in
in 0)
4-J cu "a
C TD 1
CU 1 _C
-a 3 en
o
en _i 3:
a)
cc
cr\
o
o
LA
O
1 .
CO
vO CM
CT\
cn
4-1
a
0)
E
a.
0
r.
4-J Cl)
OJ >
CU CD
s- -a
CO £=
CD
in -o
in s-
CD 3 ^
C -Q 4-J
l/l CO O
VO LA
CO O
LA VO
1 1
oa LA
r - CD
CM ro
0)
H
,
Dunty Road
Low- volume
Moderate-vo
CO
CM CO MD
- 1 Q- C
CD I s- in O
3 0 n
-------
-------
SECTION XVI '
POROUS PAVEMENT BENEFITS
Potential benefits from porous pavement, other than the savings
in cost of storm drainage facilities discussed in Section XIV, are
considered for three other areas: (1) relieving combined-sewer
overflows, (2) augmenting water supplies, and (3) providing aes-
thetic and ecological benefits.
A. COMBINED-SEWER POLLUTION
Polluted overflows from combined sewer systems currently rank as
one of the most significant and most difficult problems to solve in
controlling pollution.
Overflow pollution arises during storms when the flow in combined
sewer systems may be 100 to 200 times the amount of normal sewage flow.
Because it is not economically feasible to design intercepting sewers
and sewage treatment facilities to handle such flows, only a fraction
of the combined flow can be intercepted and brought to a plant for treat-
ment. The remainder, containing a mixture of sewage and stormwater, is
discharged directly into rivers and streams.
In most combined systems, interceptor capacity generally ranges between
only 1.5 to 4 times the average dry-weather flow, the low end of the
range being typical of most older, large cities. Consequently, over-
flows generally occur when rainfalls as low as 0.02 to 0.04 inch per
hour are recorded. Overflows have thus been known to average from 5
(Boston) to 16 (Toronto) times per month.
Although as little as 3 percent of the year's total sanitary waste-
water production may be discharged untreated in stormwater overflows,
as much as 25 to 40 percent of a year's production of suspended solids,
putrescible organic matter and bacteria will be discharged untreated as
a result of the tendency of these pollutants to settle at the bottom of
sewers only to be picked up later and discharged with stormwater.
Overflows from combined sewers have consequently made the watercourses
of numerous large cities unfit for public use, encouraged their eutrophi-
cation and rendered them aesthetically objectionable because of sludge
deposits.18
Although at present combined sewer systems tend to be outlawed in new
developments, they continue to exist in almost all of the larger cities,
particularly in the Middle Atlantic, New England, and Great Lakes regions,
and in many of their suburbs. It is estimated that approximately 18 per-
cent of the U. S. population (36 million people) is served by combined
sewers. J-"
97
-------
Present Solutions
Ignoring the need to treat storm waters, the most common method of
dealing with overflow pollution is the complete separation of sani-
tary and storm-sewer systems. For economic reasons, the usual pro-
cedure is to build separate sanitary sewer systems parallel to the
existing combined sewer lines. Such separation is becoming more
common: In the 1963 to 1967 period, 1504 miles of combined sewers 2
were separated compared to only 770 miles in the 1957 to 1962 period.
According to a recent American Public Works Association (APWA) survey,
71 percent of the 274 jurisdictions planning action to eliminate
overflow pollution intend to proceed with complete sewer separation.
Such separation is also required in all federally financed urban
renewal projects.21
Although separation is universally recognized as the most effect-
ive solution, the cost is often prohibitive. According to the APWA,
the cost of constructing a,separate parallel system averages $10,200
per surface acre, and the cost of plumbing changes to affected build-
ings averages $6,000 per acre.22 Significantly, the cost varies
widely with population density as shown below. Costs are almost pro-
hibitively high for those dense urban areas most in need of remedial
action. The following cost data have been calculated on the basis of
an average cost of $10,200 per acre for sewer separation and APWA s
survey of engineering cost studies by size of community plus the per
capita cost of plumbing changes times population density by size of
community.
Size of Community
500
1,000
5,000
10,000
25,000
50,000
100,000
500,000
999
4,999
9,999
- 24,999
- 49,999
- 99,999
- 499,999
and above
Cost per Acre
$ 1,511
2,568
3,151
4,629
6,434
6,411
12,682
39,818
Despite such extreme costs, cities such as Washington, D.C., and
Minneapolis, Minnesota, have embarked or intend to embark on complete
separation of their sewer systems.
Most of the country's major cities, however, are investigating alter-
native approaches which, although not as effective as complete separa-
tion are at least economically feasible. Such procedures include off-
system storage, screening, chemical treatment, sedimentation ponds, and
the installation of separate road sewer systems.
98
-------
Based on the relatively small sample of available data, APWA
estimates the cost of such alternative systems to average $5,100
per acre. J It is significant that the cost of such alternatives
does not appear to vary as widely with the size of community.
Porous Pavement as an Alternative Solution
The construction of a separate system of road sewers is one of
the alternative solutions to the combined-sewer problem adopted
by Toronto and being considered by such other cities as Hamilton,
Ontario and Buffalo, New York. In Toronto, overflow discharges
had, by the late 1950's made all the city's watercourses too
polluted to be used for recreational purposes. A 1960 study for
the City of Toronto recommended that a separate road sewer system
be constructed which would divert approximately 45 percent of the
city's stormwater runoff and sufficiently curtail pollution from
combined-sewer overflows to eventually render the city's water-
front suitable for public bathing. 4 The cost of this system was
estimated at $7,000 per acre (60% less than the cost of complete
sewer separation). Toronto has consequently adopted a 20-year
program of separate road sewer construction which is currently in
progress.
By its very nature, porous pavement would accomplish the same as
construction of a separate road sewer system. It would also pre-
vent runoff from roadswhich is itself increasingly being viewed
as a source of pollution^ from directly entering the watercourses.
Given the fact that city streets have an average life of about 15
years, a program of replacing existing streets with porous pavement
as their surfaces need replacement would relieve overflow from com-
bined sewers in about the same period of time as the other methods of
overflow control currently being adopted. There is thus obvious ;
potential for the use of porous pavement as a water pollution control
device.
Although the estimated cost of a separate road sewer system for Toronto
was $7,000 per acre in 1960, the value of porous pavement as a means
of.combined sewer pollution control is probably most realistically
figured at $5,100 per acrethe average cost per acre for all the
alternative solutions surveyed in the APWA study.
Based on the 19-percent road-to-land-surface ratio (including parking
lots) determined for Philadelphia,26 the imputed economic worth of
this use of porous pavement would be $5.54 per square yard for the
estimated 18 percent of the country's roads located in communities
served by combined sewer systems. The magnitude of the value of
porous pavement for this particular use clearly makes it a potential
worth considerable additional investigation.
B. WATER-SUPPLY DEFICIENCIES
Largely because the nation's population is becoming increasingly
concentrated on a small fraction of its land and is expected to
99
-------
27
double by 2020, the problem of providing adequate water to
metropolitan areas has become of national significance and some
urgency.
Inadequate water supplies are currently concentrated in two
generalized areas of the countrythe Southwest and the megalo-
politan area of the eastern seaboard extending from Massachusetts
to Virginia.
In the Southwest, moderate to severe water shortages exist in the
California, Arkansas White-Red, Texas Gulf, Rio Grande, Upper
Colorado, Lower Colorado, and Great Basin water resource regions.
Together, these regions have a total population of 44,445,000. Not
all areas within these regions, however, have water shortages, e.g.,
both northern California and eastern Texas have ample water supplies.
No precise inventory of the affected population within these regions
has been compiled; however, based on material from the Water Resources
Council, ° about 25.3 million persons or 57 percent of the total pop-
ulation of these regions reside in water-deficit areas.
Preliminary results from the Middle Atlantic Regional Water Resources
study indicate that the following urbanized areas of the eastern sea-
board have water-supply problems: " eastern Massachusetts, western
Connecticut, Rhode Island, the New York-northwestern New Jersey
consolidated metropolitan area, and the Washington, B.C., Standard
Metropolitan Area. These areas have a total population of 22,210,000.
A total then of 47.5 million persons or 24 percent of the nation's
population resides in areas of known water-supply shortages. Addition-
ally, it is recognized that many smaller communities of which no in-
ventory has been made have equivalent water-supply deficiencies.
Present Solutions
The transfer by pipeline of water from well-supplied river basins to
insufficiently supplied basins is currently the most common method
of overcoming water-supply imbalances. Such interbasin transfers are
being made for distances of from 200 to 900 miles. Based on internal
estimates, the fully amortized cost of such interbasin transfers ranges
from 30
-------
back into the soil, the use of porous pavement would increase munic-
ipal water supplies. Benefits from porous pavement depend on four
factors: (1) the average level of annual precipitation, (2) the
amount of storm-water runoff, (3) the cost of alternative methods of
augmenting water supply, and (4) the extent of the market demand for
the additional water supply that could be provided by porous pave-
ment.
In the east coast megalopolitan area where most of the affected
population is concentrated, annual precipitation for the entire
region averages approximately 42 inches. Almost all of the affected
part of the Southwest lies in the 10 to 20-inch annual precipitation
zone with an average annual rainfall of approximately 15 inches.
i
Since a certain amount of rainfall is absorbed or evaporates, the
generally accepted 80-percent runoff factor has been adopted for
assessing runoff from paved surfaces. After adjusting.for the 80-per-
cent runoff factor, each square year of porous pavement will increase
groundwater supplies by 4.52 gallons per inch of precipitation. The
42-inch average rainfall in the Middle Atlantic States would thus
yield about 190 gallons of water annually per square yard of paved
surface. With water transfers costing from 300 to 600 per thousand
gallons, the value of this increased water supply averages 8.550 per
square yard, assuming a demand exists for all the additional water
provided. After discounting at 6 percent over an average conven-
tional road service life of 18 years, the total imputed volume is 93.10
per square yard of pavement. For service lives of 6, 12.5, 15, and 26
years, the total discounted value would be 42.30, 74.10, 83.50, and
$1.12, respectively.
Similarly, the 15-inch average annual rainfall in the Southwest would
yield about 68 gallons per square yard of paved surface with an annual
value of 3.00 per year. Discounted over an average 18-year road life,
the average total value is 32.50 per square yard of paved surface area.
For service lives of 6, 12.5, 15, and 26 years, the discounted value
would be 14.80, 25.90, 29.20, and 39.00, respectively.
A market demand for additional water sufficient to justify these im-
puted economic values seems assured from both increases in population
and in per capita water use. On the basis of Philadelphia's road-to-
land-area ratio of 19 percent, national current per capita water con-
sumption and east coast precipitation rates, porous pavement would
provide an increased water supply somewhat less than 9 percent over
current demand levels. From Water Resource Council projections of
increases in per capita water consumption and population growth,
Middle Atlantic water demand should increase by 17.3 percent by 1980,
more than sufficient to create a demand for the additional water
supply. In the Southwest, which has lower precipitation and higher
rates of population growth, the comparison would be even more favorable.
101
-------
These porous pavement water-supply benefits are of considerable
importance, since existing water transfer agreements can provide
only a temporary solution to the water-supply problem. Despite
existing water transfers from the Upper Colorado, the Lower
Colorado region still has a water shortage problem. Further,
current projections, indicate that water transfers will not be
adequate to supply southern California's demands after 1990. Al-
though it is expected that desalination will be able to produce
water at 20
-------
considerable interest in color-coded surfaces for highway direc-
tional routing. By providing extremely explicit directional aid,
a color-coded routing system would decrease accidents caused by th
abrupt lane switching of drivers who had missed, misread, or mis-
interpreted posted signs. Highway color-coding has been tried in
numerous parts of the country but has been unsuccessful primarily
because of the short life of presently available color overlays
and the lack of a uniform national color code. If a durable
colored surface could be developed it would doubtless have a con-
siderable market among state highway departments.
Aesthetic Enchancement
Aesthetic considerations dictate a demand for color infusion in
three types of pavement: residential streets, specialized re-
creation areas, and parks.
Conversations with a number of developers and urban designers
indicate a preference for paving residential streets in quiet
earth tones which would blend with the natural environment and
provide a more restful setting than is currently possible with
the starkly unnatural contrast provided by blacktop. It is
evident that there would be considerable demand for such color-
infused surfaces if they could be produced at a competitive cost.
Park officials expressed an equal enthusiasm for using color-
infused asphalt surfaces. Officials of the-National Park
Service were especially interested in such surfaces because they
are committed to attempting to eliminate blacktop roads from the
National Parks and have actively tested various color overlays
without finding any of them sufficiently durable.
Acrylic color overlays are currently used in specialized rec-
reation areas, such as tennis courts where surface wear is not
a problem, to provide a more attractive surface. Color-infused
porous pavement would be desired for aesthetic reasons if it
could be produced at a competitive price.
Temperature Control
The use of colored asphalt for temperature control was judged
by urban planners as being particularly valuable for use in
dense residential areas, particularly inner-city urban-renewal
districts. The placement of light tan pavements in urban-
renewal areas would eliminate the unpleasant heat absorption
effects of blacktop surfaces as well as provide a cooler and
more restful appearance.
Removal of Curbing
The elimination of curbing was initially expected to be one of
the major aesthetic benefits to be obtained from porous pavement.
103
-------
The opinions expressed by developers and architects on the aes-
thetic value of eliminating curbing were, however, rather mixed.
Some saw it as an aesthetic improvement and others viewed it as
a rather neutral development, possibly because they have become
so accustomed to viewing it as an integral part of residential
developments. In almost all cases, it was felt that public safety
dictated that curbing should only be eliminated in low-speed, low-
traffic residential areas.
Despite the fact that no clear-cut aesthetic benefit was found, the
elimination of the necessity for curbs and gutters to channel water
to storm inlets would provide a considerable cost savings of from
$2.50 to $4.00 per linear foot of building lot frontage.
Specialized Surfaces
As a result of our analysis of the cost of conventional pavement
surfaces, it was decided that, in most instances, exotic porous
paving surfaces, such as perforated butyl rubber sheets and modi-
fied carpeting, would not be economically feasible. Only one
area of significant potential for such surfaces was evidenced
safety areas under playground apparatus. Specialized materials,
such as woodchips, rubberized materials, and special carpetings,
have been used experimentally in such areas. The cost of these
experimental materials has sometimes exceeded $10 per square yard.
However, prevailing opinion indicates that a cost no higher than
$5 to $6 per square yard and relatively good wear characteristics
would be necessary for general use in safety areas under playground
apparatus. In general, it is questionable whether the limited
market demand for exotic surfacing materials would justify their
development and production.
Ecological Benefits
Several ecological benefits were identified primarily involving
relief from flash flooding and the preservation of normal drainage
patterns as well as, to a more limited extent, the preservation of
roadside vegetation.
Relief From Flash Flooding
In outlying suburban areas, storm runoff from large paved surfaces,
primarily parking areas, is outflowed through storm drains directly
into the nearest stream or other adjacent watercourse. This practice
produces considerable flash flooding and stream-bed erosion during
periods of heavy rainfall. Demand for porous pavement to eliminate
flash flooding is probably limited at present since most flash flooding
results from private developments. In some areas, for example in Spring-
field Township, Pennsylvania, action has been taken to protect streams
by forcing developers to control release of runoff. Greater restric-
tions are anticipated from increased public ecological awareness and
should create a natural market for porous pavement.
104
-------
Preservation of Natural Drainage Patterns
Impervious paved surface have the detrimental effect of interrupt-
ing natural drainage patterns. Disturbance of natural drainage
patterns is undesirable where paving is imposed on otherwise open.
areas whose natural character is worth preserving and/or areas
where the surfacing will only be temporary. Our investigations
indicate that quasi-public organizations which have large areas of
parking on scenic open land are actively searching for paving sur-
faces which would preserve the natural drainage pattern as well as
prevent spoliation of streams through flash flooding. Additionally,
representatives of the City of Philadelphia have expressed con-
siderable interest in the use of porous pavement for the parking
areas at the Bicentennial Exposition so that the land can later be
returned to its original state. It is anticipated, of course, that
the market for porous pavement for this use would be largely limited
to public and quasi-public authorities.
Preservation of Roadside Vegetation
Only very limited demand was evidenced for the use of porous
pavement to prevent water starvation of roadside vegetation.
It was generally agreed that roadside vegetation does not suffer
water starvation in single-family, residential, and other low-
density developments. The only areas where it was generally
felt that roadside vegetation was sufficiently water-starved for
porous pavement to be beneficial were urban parks in large cities,
paved areas in hi-rise apartment clusters, dense urban areas where
little exposed ground exists, and large paved surfaces where plant-
ings may be desired.
Economic Evaluation of Aesthetic and Ecological Benefits
In most cases it has not been possible to rigorously establish
the economic value of the aesthetic and ecological benefits
identified since alternatives to obtaining' these benefits from
other sources are not currently in use. Accordingly, the govern-
ment and private officials whc identified the various benefits were
asked to evaluate them in terms of the premium they would pay to
obtain them. In most cases, the market value placed on the benefits
was relatively modest. Typical premiums estimated for the various
benefits are enumerated in Table 16. In some cases, values were not
assigned because the particular benefit would require governmental
and/or legislative action to create a market demand. In general,
premiums for certain specialized uses such as parking lot and play-
ground color infusion are relatively high, whereas for more general
uses, such as color-infused residential streets, the assessed value
of the benefits is low in relation to the total pavement cost.
105
-------
I
5-
01
(O
^
u_
LU
z
LU
CO
IETIC AND ECOLOGICAL
ZC
LU
e£
IT
1
5
LU
0
14
O
z.
o
o
LU
U
3
O
|
O
U
111
Type of Pavement
.t!
u
c
m
CD
o
a- ^_ . 0
ro 4-1 CM TJ 4J
°- § ^ c -5 in
4J L- CM O Ol C
C 1- >- LA ~ T3 O
5 3 -C in 0)
u O O ra cM> . 4-> CM > 4J > O 11)
OOOCM CM _ o m
. . -co- w -
4-j ro > ro ^~*> c "
Q) m u in i- en a)
nl L.IDin* c
-o Nino "~C,4J,,L
4J ro .-m cm in 3 - in 1-
co o in ro 4-1 ^ >. DiO O4->
mi- roro -QdJmro c L. i- u c
-Q M-c:i-D UJ1-J3
mi- rocn L. ro c 4-> u
G)ID Q. in 4-> 3 m » ro ctn3
£3CQ.l/) CO DCQIQ--^
(-)
2
4-J .
"o o
U O
4-J
§ 2
0000 3
4-1 4-j 4-j o ro
Q) (U 1) U
4J 4-1 4-1 M- Q.
in in i/i ID E=
u _u ^ ^ ,£
§
U
(/)
3
**
C
1
o
Q
O
CM
O
-a-
CM
O
4->
O
LA
^
in
Residential street:
o
Aesthet
_.
_
_
_Q
1
^
O
(4
Q
(13
i
0)
OC
a
T3 0)
c -t-1
a .
= E
D
. ~
* ^^
^ ct" eo
-a 4-j a) CM
u ro > >
5> in c o
_ . o
in en
in a) CM
ro >-
m a) CM o
3 .a 4J
ro 4-j o o
> in 4-1 in LA
3 (U
O E in 0
Z. 3
in in
0) (0
O O C
ID ID ID
14- 4- J3
l_ 1- l-
333
in in
.* J^ TJ
1_ L_
ID ro c
Q. Q.
in
QJ QJ m ro
en co ~o a)
u u ID L.
ro ID o ID
_t _i cfi
c
o
-C 0) 4J
in en ro
"o c o a)
14- en
c ro cm
E 0 i- 0 >
o -a
u en 4-j 4-» aj
if- c ro ro tj
._ > ro >
14- -a L_ L. i- m
-
cn c. -t-1
* ID
t/'
«rj* _Q
t L. ro
o ^
J ro
(U L, >
a m
?4 CO TJ
o ~o o
to -D . m
. S; O 1-
w^-o x
i_ g5 m m
St. 5 m
° s-i £
m K O a)
-ri o in
c. i3> <1>
c s; L. i-
m S o a.
4-1 U-
u. O
1_ QJ 4J 4-1
0 Q. 0
<4- O T
in 't- m
m >- i-
Sro L.
ro in
ro L. m .-
3 O fl
" L. L. -
O O tt) I-1
._ a.
">. o o m
u o >-
o m
ro ro -=r -s
TJ C
,. i 4-J l/V
ro i- o m 3
4-j m LA a)
in a. i- TJJ
c x CM 4J m
. m -to- in 4->
'o 'o o "ro ~
in in in c 5"
O O O m i-
O O O T> 4)
4-1 4-J 4-J in
c c c m a)
m m m L. in
I- U U O
u. i- u m -c
33 3 T> 4.J
o o o -
c c c ? -:
o o o *->
o l-
o ~a -a o
-------
It was repeatedly emphasized throughout this study, however, that
in order for these premiums to be sustainable, it is absolutely
essential that the pavement require no more maintenance effort than
current surfaces. As most public officials observed, it's a lot
easier to get money to build roads than to maintain them. It is
precisely these increased maintenance costs that have prevented
colored asphalt overlays from coming into general use.
Additionally, particularly in the residential sector, a signif-
icant extra premium could be obtained if a porous pavement material
could be designed capable of being laid during the winter. Current
asphalt surfaces must be laid when -the temperature of the surface
is 50°F or higher. This limitation means that road surfaces must
be laid during the summer and that most fully developed building
lots thus have to be carried to the following spring/summer -selling
season. Carrying costs of $250 to $400 per building lot, or $1 to
$1.25 per square yard of road surface, would be eliminated if road
surfaces could be laid during the winter.
107
-------
-------
SECTION XVII
MARKET DEMAND FOR POROUS PAVEMENT
The market demand for porous pavement has been assessed in three
phases: (1) the total annual pavement demand for the types of
pavement under consideration has been estimated, (2) the cost of
porous pavements of equivalent design has been assessed and (3) the
total annual pavement demand, cost of conventional pavement surfaces,
imputed premiums for specialized porous pavement benefits, and the
cost of porous pavements for designs equivalent to conventional sur-
faces have been related to determine the potential market demand for
porous pavement at various price levels.
A. CONVENTIONAL PAVEMENT DEMAND
Estimates of conventional pavement demand by pavement type have been
derived from a number of primary and secondary sources. An attempt
has been made to arrive at as comprehensive estimates as possible;
however, because of the lack of comprehensive data sources, it has
not been possible to include minor sources of demand in all categories.
Derivation of demand estimates for each pavement type is discussed in
the paragraphs below, and these estimates are summarized in Table 17.
Parking Lots
Pavement demand estimates for parking lots have been based on the
three major users of surface parking lots: shopping centers, office
buildings and industrial plants. On the basis of the National Park-
ing Association's estimate that 500 shopping centers are constructed
annually with an average parking area of 50,000 square yards, the
annual demand for shopping center parking surfaces has been estimated
at 25 million square yards.
Demand for industrial plant parking lots has been estimated to average
11.6 million square yards per year on the basis of the annual increase
in manufacturing employment, 1960-1969, and an Institute of Traffic
Engineers' survey of the ratio of parking spaces to number of employees
in manufacturing industries.30
Demand for office parking lots has been estimated on the basis of new
suburban office parking demand since most city office buildings rely
on parking structures rather than street-level parking. Demand for
suburban office parking spaces is estimated to grow by 1.45 million
in the next six years,31 implying an annual street-level parking demand
of approximately 9.4 million square yards.
Total annual demand for large parking areas, based on the three major
use categories, was thus estimated at about 46 million square yards.
109
-------
Residential Streets
The total volume of residential street construction has been esti-
mated on the basis of the annual level of single-family housing
construction and the average lot frontage of single-family dwellings.
On the basis of recent statistics, total housing starts for 1971 are
estimated at 1.8 million. Of these units, 59.7 percent are planned
as singlefamily residences and approximately 15 percent as town-
houses. 32 L0t front footage was estimated at 90 feet for detached
O O
single-family residences and 31.6 feet for townhouses. ° Assuming
a standard 30-foot-wide residential street, this level of housing
construction implies construction of 8270 miles of residential roads
or approximately 145.6 million square yards of residential street
surfaces.
The total estimated residential street construction has been segre-
gated into demand by design level on the basis of the location of
construction: non-SMSA (Standard Metropolitan,Statistical Area),
suburban county within SMSA, and metropolitan city.
Recent housing construction data were available for 40 of the
country*s 115 major SMSA's. On the basis of extrapolation by
population, 68.5 percent of residential construction is estimated
to be within SMSA's and 31.5 percent in non-SMSA's.34 Based on
Philadelphia's housing data, 12.1 percent of SMSA construction is
estimated to be located within central cities. Annual housing starts
are thus estimated as follows: 8.33 percent in metropolitan areas,
60.14 percent in suburban SMSA counties and 31.53 percent in non-SMSA
counties.
These statistics imply annual residential road construction of 689
miles in metropolitan cities where high design standards prevail,
4974 miles in suburban SMSA counties where the upper end of the low-
design range (8 or more inches total thickness) is generally observed
and 2607 miles in non-SMSA counties where the lower level of low-
design construction (less than 8 inches total thickness) is generally
observed.
Business Streets
Estimates of the annual volume of business street construction were
based on the standard developers' total road construction break-
down of 85 percent residential streets and 15 percent business streets.
Based on this ratio, a total annual construction of 1458.7 miles of
business streets associated with single-family residences and 984.6
miles associated with apartment dwelling units was estimated. Of this
total, 8.33 percent, or 203.6 miles, was estimated to be city business
streets on the basis of the construction location data given in the
previous sections. Assuming a standard 40-foot-wide business road, a
total demand of 57.3 million square yards was estimated: 4.8 million
in metropolitan areas and 52.5 million outside metropolitan areas.
110
-------
County Roads and Highways
Data on the total level of county road and highway construction were
obtained for 1969 from the U.S. Department of Transportation. 5 In
most instances, the distribution of average daily traffic (ADT) vol-
umes was not available from which to segregate total construction
into the four general ADT classes identified in this study. Con-
struction estimates for these four road classifications were there-
fore interpolated using data on the design level of existing roads^5
and the ADT data that was available for the primary system. In mak-
ing these estimates, the following road systems were considered:
rural primary and secondary roads, secondary and primary municipal
extensions, interstate highways and local and forest roads.
Low-Volume County Roads
Low-volume road construction on the primary road system was estimated
on the basis of ADT data for existing primary mileage. Low-volume
road construction on other road systems was estimated by assigning a
percent of new construction to this sector equal to the existing per-
cent of low-design roads on each system.
Moderate-Volume County Roads
The annual construction of moderate-volume roads was estimated from
total road construction on the basis of the percent of roads with ADT
volumes of 800 to 2000 on the primary road system. For other road
systems, this ratio was adjusted on the basis of the variation in the
distribution of design levels of existing roads for each road system.
Highways
The remaining constructed mileage of moderately-high-level-design and
high-level-design roads was segregated between two and four-lane high-
ways on the basis of the distribution of roads by width on the primary
system. For other road systems, adjustments to this ratio were inter-
polated on the basis of differences in the distribution of moderately
high-level-design and high-level-design roads for each road system.
From the above data analysis, the following annual road-construction
estimates were derived:
Low-volume county roads
Moderate-volume county roads
Two-lane highways
Four-lane highways
15,009 miles
14,479 miles
10,082 miles
3,694 miles
Based on average widths of 22 feet for low-volume roads, 24 feet for
moderate-volume roads and two-lane highways, and 48 feet for fpur- .
lane highways, the following annual pavement demand levels were
estimated:
111
-------
Low-volume county roads
Moderate-volume county roads
Two-lane highways
Four-lane highways
Sidewalks
193.6 million square yards
203.9 million square yards
141.9 million square yards
104.0 million square yards
Sidewalks are primarily provided with residential streets, business
streets and the municipal extensions of primary and secondary high-
ways. Total annual construction of these, roads (based on data from
previous sections and Reference35 is estimated at 14,113 miles, im-
plying an annual construction of 28,226 miles of sidewalks or 66.9
million square yards of standard 4-foot-wide sidewalks.
Playgrounds
Paved playground areas are generally constructed with or associated
with school construction. Paved playground areas recommended in a
1969 School District of Philadelphia study36 are 0.24 acres for
elementary schools and 0.68 acres for junior and senior high schools.
Elementary and secondary school construction estimated on the basis
of pupil growth from 1968 to 1970 and the average number of pupils
per school37 indicates an annual demand for paved playgrounds of 833
acres or approximately 4.4 million square yards.
Public Squares
Relatively precise data from which to estimate public square construc-
tion was not available. Accordingly, we assumed a construction level
of one public square-per year for each of the country's 115 major
metropolitan areas and for each of the approximate annual total of 40
new residential developments exceeding 1000 homes. Assuming a rep-
resentative plaza size of 15.5 thousand-square yards, the total pave-
ment demand would be approximately 2.4 million square yards per year.
Airport Runways
Based on data from the Portland Cement Association, it was estimated
that major private airport runway construction (FAA continental and
intercontinental standards) averages about nine runways per year.
A total pavement demand of approximately 1.5 million square yard is
thus indicated for a standard runway measuring 10,000 feet in length
and averaging 150 feet in width. Total runway construction was
estimated to be approximately 25 percent asphalt and 75 percent concrete.
Demand Summary
Total annual demand for pavement surfaces, estimated as above, for
each pavement category are summarized in Table 17.
112
-------
Table 17
PAVEMENT DEMAND ESTIMATES
Type of Pavement
Parking Lot
Residential Street-
Low-design: SMSA
Non-SMSA
H i gh-des ign
Business Street
Suburban development
Cityf
County Road
Low-vol ume
Mode ra te- vo 1 ume
Highway
Two- lane
Four-]ane
S i dewa 1 k
Publ i c Square
Playground
Major Airport Runway
Mi les
4,974
2,607
689
2,240
20k
15,009
14,479
10,082
3,694
Surface Area
(Thousand Sq. Yds.)
45,986
87,538
45,887
12,129
52,547
4,776
193,628
203,864
141,955
104,023
66,926
2,400
4,397
1,500|
Width assumptions used in making these calculations are residential
street, 30 ft.; business street, 40 ft.; low-volume road, 22 ft.;
moderate-volume road, 24 ft.; two-lane highway, 24 ft.; and four-
lane highway, 48 ft.
t Includes only city business streets associated with expanded resi-
dential/commercial development. Major city traffic arteries are
included under highways.
t Estimated construction breakdown: asphalt, 25%; concrete 75%.
113
-------
B. COMPARISON OF CONVENTIONAL PAVEMENT WITH POROUS PAVEMENT
Porous pavements have been designed for various levels of traffic
volume on the basis of Asphalt Institute of America and California
State Highway Department specifications. In establishing these
designs, two basic assumptions have been adopted: (1) design thick-
nesses are based on the assumption of a poor subgrade to allow for
the low load-bearing capacity of wet subgrades and (2) a minimum
16-inch crushed-stone base has been adopted in all pavement designs
to provide sufficient base reservoir capacity to handle a 5.4-inch
maximum daily precipitation. On non-clay soils (permeability s 0.04),
a maximum of ten days would be required for this level of precipita-
tion to percolate into the subsurface soil.
Porous pavements designed to the above specifications are classified
in Table 19 on the basis of design traffic number (DTN): DTN values
below 100 represent low and moderate traffic volumes; DTN readings
between 100 and 5000 indicate heavy traffic flows. A complete descrip-
tion of the derivation of porous pavement design by DTN classification
is given in the portion of this study on physical design.
Table 18
POROUS PAVEMENT DESIGN BY DESIGN TRAFFIC NUMBER
Design Traffic Number
1
10
20
50
100
1000
5000
Crushed-Stone Base
(inches)
16
16
16
16
16
20
22
Asphalt Concrete
Surface (inches)
4
A
H
5
5
6
7
These DTN classifications have been used to establish equivalent porous
pavement designs for each of the conventional types of pavement con-
sidered in preceding sections. Table 19 gives the DTN classifications
and equivalent porous design for each of the major types of road. In
constructing this table, several exceptions to the original pavement
classification have been made. Public squares and sidewalks have been
omitted because the surface qualities of porous asphalt concrete were
not considered aesthetically suitable in this use. Because of the
relatively small difference in porous design, low-volume and moderate-
volume county roads have been combined. Further, airport runways have
not been considered in this comparison because the design specifications
adopted could not be used to establish porous runway designs.
114
-------
Table 19
EQUIVALENT POROUS PAVEMENT DESIGN
Type of Pavement
Parking Lot
Residential Street
Low-des i gn
High-des ign
Business Street
Suburban development
City
County Road
H i ghway
Two- lane
Four-lane
Playground
DTN
1-10
1-10
10-20
20-100
100-1000
10-100
100-1000
000-5000
1
Crushed-Stone Base
( inches)
16
16
16
-------
UJ
CD
Q
S.
O
O
CM
d)
S
UJ
3E
UJ
C5
Ul
O
O
D-
D;
o
a_
o
to
HM
a:
S5
co
o
o
c
<0
C
>
c
o
o
c
-
ID *^s
L, -CO-
*
4-1 \
tn -co-
0 «
o
4J
10
o
0) CM~*
CF> >-
u -co-
o) ~-^
^
<0
en
c:
ro ,*
QC CM
>^
4-" \
tn />
O ^-'
o
+-1
c
(U
fO
u-
o
tO
Q.
H-
CO
-3-
,^
oo
La
i
La
vD
vD
VD
K
6
,
vO
1
CM
.
vO
4_j
o
O)
c
V*
l_
ID
D_
O CO
CM CM
r^ o
o
o
vD
QT\ t
1 1
CD La
OO CO
^l- oo
Laoo
CO CO
VD vD
K
co La
vO vD
1 1
co en
CO
» ,
VD VO
4J
a)
a)
u
4-1 C
co c en
CO
, tn
CD tn a)
tu ~a
4-1 -a i
C 1 -C
a) 3 en
-a o
1 3=
tn
a)
0£
O 0
-^- CM
CO OO
.-
o
O CM
o-^
CM
1 1
CD O
en CM
VD -^J"
00 VO
COvD
VD r~-
cn CM
en co
vO CO
1 1
La en
vo en
vD vO
C
a.
o
*
(D
(U C
co ns
-Q
in u
in 3 >-
(U -Q 4->
d 3 *~
~ CO CJ
tn
3
CQ
O
CM
cn
CO
CM
CO
1
1
-y
CM
^
vD
r*%,
CO
vD
1
^~
O
vD
-a
to
o
>^
4-1
C
Z3
o
o
O OO CO
La oo oo
^r co La
CM CM
La o o
-3" -^" cn
O CO
CM CO CO
1 11
-3" vb vO
La co r~~.
* * *
oo vo r-.
VD oo -=r
co -d- vo
r-^ co oo
CM cn-=r
cn r~» o
i^ co cn
i i i
O vo CO
CO CM
vD OO CO
0)
0) 4-< 4-1
(U C (D
C (D flj >~
(0 -C O
^-i a. c
>« i L, tn O
to O 3 < <->
220
-C I U-
cn
3:
La
La
^j-
o
CM
La
i
o
cn
CO
vD
CO
-3-
^
vb
CO
-^"
t
LTV
CO
-=r
"O
c
3
O
i
CO
^.
fD
r
a.
o
o
o
10
to
o
X
0)
o
0)
(U
14-
a
in
4-1
o
cu
ii
14-
0)
D)
C
(D
116
-------
CXI
4| ,
fK
0
o
^
Q.
Q.
3
in
u
0
as
3;
0)
Q)
CO
1 4-
0 ID
C
_Q ryf
E
O
O
<1>
3
CM .
73 -»
Q) -
-0
T3
"O Q)
U! C -I-1
Ul (13 O
O E CD--J
O Q t- £
(1)
3
"(0
> .^
CM
X- >-
0) ~-~.
-a -t/v
< "^
TJ
-0 CM
a >s..
a) {/)-
-a -
TJ
~jj
-a
L. 0) -^
O > "»-
n_ ra * '
CL.
(U
ra oi
C +J (TJ
O C C
w E ra -'
C HI L. CM
a) > Q >.
> to \
c o_ _c
o
3
T3 CM
ro c >-.
3 CO
CEO
C a> 0
L- a)
4-* O
CO
c
U> 1)
Ul -Q
C 3 >.
Ul 3
3 , CO O
m
LA
I**-
^J
1
O
Ul
O
LA
V
01-3-
CO CO
O O
1 1
en 01
CM 01
o o
^f «^f
CM CM
LA LA
*
LA LA
CO
-3"
-3-
VO
0
CM
en
co -3-
CM \O
VOCO
CA Ol
en o
CM
*fl)
c
o
-a a) >
ID E 1
O 3d)
a: 4-<
O ro
4J i m
C 3 T3
3 O O
O -J Z
0
o
LA
CA
1
O
LA
CM
^
1
LA
u_>«
-3- -3- CM
CO CO
O O
1 1 1
01 01 en
CO Ol Ol
o o o
-3" -3" -3"
CM CM CM
* -3" -3-
LA LA LA
LA LA LA
OO CO CO
VO CO -3"
01 -3- \r>
r~. co co
o oo 01
LA OO CO
^J- 01 LA
-~ CM CM
LA Ol
LA CM
cn o
-3"
-3- O
0) ^ ra ra L.
ra i JCT o
3 I L- Q. C
jn o 3 ui o
01 3 o< o
. K- Lu
3:
o
CM
CM
1
0
0
^~
o
CO
a-
01
\o
LA
LA
a-
f^.
en
01
i
-3-
o
c
3
O
L.
Ol
ra
Q-
117
-------
o
O
CM
n
V. Cl
9} O
flj ^
a.
o
4-1 M-
ra CM
o -.
0)
"O '
a
**
-o
o u
in c 4-"
1/1 ro o ^~
3 E a) *e
L. Q) M- ' '
CD Q M-
(D
3
10 f~^
> CM
a "v.
a)
o ^^
T3
<
T3 4)
Ul C »J
tn ID O *-~*
O E O ^9
3
IB
CM
-o >-
4) .
a xrt-
-o
o
a a>
ro o "S-9
0) «-
o «-
^
4-1
C
e
§
ro
a.
M-
O
4J
C ID 01
>. 10 i ID L.
ra i J: o
Jl 1. O. C
o 3 in c
en 30^0
a:
[ayground
a.
118
-------
On a strict cost-comparison basis, which does not take into account
any of its environmental and safety benefits, porous pavement has
the potential in all uses, except parking lots, low-design residen-
tial streets and playgrounds, of eventually approaching a market
equal to the indicated gross annual demand subject to the following
qualifications:
1. The pavement designs from which these cost comparisons have
been made assume construction over non-clay soils (perme-
meability _> 0.04). The potential market demand will be
reduced to the extent that road construction occurs over
clay soils although no reliable figures are available upon
which to estimate this extent. Test borings will be re-
quired as each road site to evaluate the soil's suitability
for porous pavement.
2. Cost comparisons have been made on the basis of average
Middle Atlantic precipitation. In areas of very high pre-
cipitation, a deeper, more expensive reservoir capacity
would be required unless the subgrade has a very high per-
colation rate. Correspondingly, in areas of low precipitation
and/or high percolation, a cheaper porous design could be used.
3. For these cost comparisons to be valid, porous pavement main-
tenance costs must not significantly exceed those of conven-
tional pavements. Since porous pavement design uses current
road construction materials (asphalt concrete and crushed
stone), it has been assumed that maintenance costs would
approximate present levels. For porous pavement to realize
its market potential, satisfactory maintenance experience
would have to be experimentally demonstrated.
4. The potential indicated could only occur over 'a substantial
period of time as the feasibility of porous pavement is
demonstrated in use. No basis exists from which to satis-
factorily gauge the speed of acceptance.
For the remaining pavement types the potential market is smaller.
Porous pavement parking lots and playgrounds are only cost competitive
where soil or precipitation conditions are more favorable than those
conditions on which our cost comparisons were based. Similarly, on
the basis of our average cost range, porous pavement would be cost
competitive for only about one-half of the low-level-design residential
streets. In many cases, however, porous pavement would be economically
feasible in these uses on the basis of the value of additional benefits
obtainable from porous pavements. These benefits and their markets are
discussed below.
The value of applicable porous pavements benefits and their relative
demand are indicated in Table 21. These benefits are enumerated
separately although, in many cases, part of the demand for one benefit
119
-------
will intersect with the demand for other benefits, implying a
correspondingly greater total benefit premium for porous pave-
ment.
The most significant benefit identified is the relief from combined-
setter pollution. As shown in Section XVI, this benefit would add
an estimated value of $5.54 per square yard affecting approximately
18 percent of the total pavement market. The imputed value of this
benefit is not, however, realizable in the private market, but is
realistic only if a public program were initiated to utilize porous
pavement to relieve combined-sewer pollution.
Based on the present distribution of water-supply deficiencies,
porous pavement could command a premium value as a means of water-
supply augmentation in 24 percent of the total market for paved
roads and parking surfaces. The imputed premium for this benefit
varies from $0.15 to $1.12 per square yard of surface area, depend-
ing on the service life of the paved structure and the level of
precipitation in the area of installation.
The demand for colored parking surfaces is estimated at 20 percent
of the total parking lot market, equal to the percent of the total
parking area demand represented by street-level office parking lots
where the demand for colored surfaces is extremely strong. Accord-
ing to industry representatives, a variegated colored surface would
command a premium of $1.00 per square yard.
It is estimated that 40 percent of the market for low-design residen-
tial streets would consist of colored surfaces if they could be
obtained at a 10-percent premium over present costs, or about 30
-------
public highways. If a suitable colored material were found,- it
could eventually be expected to garner 5 to 10 percent of the
major highway market. A premium of $2.25 to $3.50 is estimated
on the basis of the relatively unsatisfactory colored materials
currently in use on highways. A premium of $0.50 to $1.75 per
square yard would be commanded by colored pavement in the low-
volume county road market. This premium would apply to only that
small friction (<5%) of this market which consists of park roads.
The strongest and most urgent demand for colored surfaces was
evidenced in the paved playground market. Most public recreation
officials are keenly interested in obtaining durable and economical
colored surfaces. A market premium of $1.00 to $2.00 per square
yard of colored surface, depending on the'type of playground and its
location, was indicated for 80 percent of the playground market.
Demand for porous pavement to relieve flash flooding and preserve
natural drainage patterns is currently limited to less than 5 per-
cent of the total market for parking surfaces. In this limited
use, such an ecological benefit would command a premium of $1.00 to
$2.00 per square yard. If, as anticipated, increasing restrictions
are placed on the disposal of stormwater from large parking surfaces,
the market demand for this benefit of porous pavement would increase
substantially. A premium value of $0.50 to $1.00 was placed on the
use of porous pavement to preserve roadside vegetation. As discussed
in Section XVI, the demand for this benefit should affect only a very
minor proportion (<5%) of the parking, low-level-design residential
street and county road markets. In the city residential street market
where .vegetation is starved by extensive paved surfaces, the demand
would vary from 10 to 25 percent of the market depending on the density
of development.
Because cost economies range from $1.50 to $2.40 per square yard, a
demand for porous pavement to eliminate the need for curbing can be
expected in 85 percent of the suburban residential street market and
in 50 percent of the city residential street market. The anticipated
demand is less in the city because of the substantially greater number
of high traffic residential streets where the elimination of curbing
might cause safety problems.
One item omitted from Table 21, which remains to be investigated, is
the economic benefit of using porous pavements as a potential means
for reducing the capacity of and flows to storm and combined sewer
overflow treatment facilities (to be required in the near future).
The savings due to the reduced cost of smaller treatment facilities
and their reduced operational and maintenance costs could redound
significantly to the benefit of porous pavement.
121
-------
As shown in Table 21 the range of premium-commanding .benefits
offered by porous pavement is quite substantial. These benefits
can be expected to render porous pavement economically feasible
in those low-level-design uses (parking lots and playgrounds) where
it is not generally cost competitive with conventional pavements,
and to increase the economic advantage of porous pavement over
conventional pavement in high-level-design installations.
122
-------
SECTION XVIII
ADDITIONAL TYPES OF POROUS PAVEMENT
A. PORTLAND CEMENT CONCRETE STRUCTURES
Design
The object of the design study conducted for cast concrete pavement
was to provide a relatively heavy duty pavement which would trans-
mit water to the subsoil and which would store water beneath the
pavement surface until percolation removed it. The appraoch taken
was to design a pavement having a surface through which holes could
transmit the water and a supporting structure which would also
serve as a storage reservoir.
In order to avoid the use of a thick surface which would support
the load but which would leave little room for reservoir capacity,
it was decided that the reservoir structure beneath the surface
should also serve as the main load bearing member. This considera-
tion dictated that a web, or grid-like, structure be adopted and a
trade off be made between web thickness, depth, and reservoir capac-
ity. In addition, the size of the reservoir openings was limited
by the safe span for the surface, or cover, when under load.
For test purposes a web and cover pavement was prepared. The pave-
ment, shown in Figure 13, consists of the supporting and the surface
cover. The web openings were 5x5 inches square and 4.5 inches
deep, with a web thickness of 2.0 inches. The cover was 2.5 inches
thick and contained .75 inch diameter holes which were situated over
the centers of the web openings when the pavement was assembled.
Test Apparatus
In order to test the assembled pavement, the apparatus shown in
Figure 14 was used. The apparatus consisted of a loading jig, test
tank, and load/deflection transducers with recorder.
The loading jig was designed so that the test tank could be loaded'
into the jig in its entirety. Load was applied to the test pavement
in the tank by means of a 5 inch diameter hydraulic cylinder, equip-
ped with a load pad of 35 square inches.
Test Tank_
The test tanks were designed for dual purposes: First, for use in
load testing of pavements, and, second, for use in determining the
extent of bacteriological activity beneath the pavements. The tanks
were designed with slanted bottoms to collect percolated water, and
with steel grid false bottoms to support the pavement and subsoil.
123
-------
I 11*15
Figure 13. Web and Cover Pavement
124
-------
+J
ta
i-<
ctf
03
0)
H
I
I
PW
X)
0)
0)
CO
to
rH
(U
M
t>0
H
125
-------
A three inch layer of coarse gravel followed by a two inch layer
of coarse sand was placed on the grid and then subsoil of the de-
sired thickness was added. The pavement and its base course were
then applied.
Load/Deflection Transducers
Pavement deflection was monitored by using a spring loaded, linear
potentiometer. The transducer was capable of resolving deflections
of .001 inch. A load cell with 10,000 pounds capacity was mounted
between the hydraulic ram and the load pad. The load cell read-out
was capable of resolving 100 pounds load at the recorder setting
used. The signal from the linear potentiometer was displaced on
the X axis of an X/Y recorder while the load cell signal was dis-:.
played on the Y axis. A plot of the stress/strain curve was thus
obtained.
Experimental Results
Tests with the experimental pavement were initially satisfactory
but after load cycling under wet and dry conditions, cracks devel-
oped in both the cover and the supporting pattern. Loads from 0
to 6000 pounds were applied to the 35 square inch load pad in load
times of 10 seconds and shock load. After each loading cycle when
load was returned to zero a permanent deflection was noted. The
amount of the permanent deflection per cycle decreased during dry
cycling but when the pavement was wetted the deflection rose and
then decreased again with successive cycles. Before the new con-
stant value was reached, the pavement cracked. Two pavements were
tested and the same crack patterns developed.
Examination of the two pavements indicated that the reason for fail-
ure was shifting and settling of the subgrade under the load applica-
tion point. This led to a relatively large unsupported span at the
maximum chord of the supporting web and the failure so induced in
the web led to failure of the cover.
The use of a concrete pavement of the type tested is not feasible
unless subsoil and base shifting can be eliminated. Further work
would be required to determine whether a gravel course lightly bonded
with asphalt would have the requisite stability.
B. FABRICATED PAVEMENTS
The following section of the Proposal describes approaches to porous
pavements made up of factory-fabricated components, to be assembled
at the road site. These were not investigated further after it was
found that a less expensive open-graded asphalt concrete should be
suitable.
126
-------
Bricks
Pre-formed bricks, or other components which are designed to main-
tain large cracks between them, are laid on compacted graded gravel
and the chinks between loosely filled with fine gravel (see Figure
15). Vermiculite, perlite, or other very porous minerals may be
better than gravel for this purpose, and should be investigated.
The bricks may be shaped or textured on the wear surface to promote
water run-off and to provide skid resistance. Materials to be in-
vestigated for the bricks include modified concretes and other min-
eral compositions, as well as ceramic clays. The supply of the
bricks in large slabs, and the handling of these by machine, is in-
dicated to keep installation labor costs down.
As shown in Figure 15, each brick is capable of a small vertical
displacement under load, and the resulting changes in void volume
in the underlying gravel in dry weather should pump- air into the
substrate with each passage of a vehicle. An alternative arrange-
ment (Figure 16) might be to put two holes through each brick and
use rods or cords to tie large areas of brick together. In this
case, spacing between the bricks could be maintained by spacers
such as thick washers, if not by humps in the bricks themselves.
The relative merits of this configuration, in terms of mechanical
properties and air pumping, remain to be determined.
O
Figure 15. Bricks with Lugs to Control Spacing Between (Shaded
area loosely packed with gravel)
127
-------
Figure 16. Bricks Assembled on Rods with Spacers Between
The first alternative has the advantage that the materials can^be
conventional - brick pavings have a long history - and their life
and ease of repair are both known and favorable.
Honeycombs
The wearing surface is a perforated butyl rubber sheet about 1/8
inch thick. This covers a honeycomb structure which in turn rests on
gravel (see Figure 17). In this configuration, the rubber sheet
is deformed with each passage of a wheel, hence acts as an air
pump. On a sandy loan soil or other withigobd load bearing proper-
ties when wet, the honeycomb could rest on strips or a sheet con-
taining large openings, thus dispensing with a gravel subbase (see
Figure 18). The reservoir capacity of this structure is within
the honeycomb itself and can be quite large.
The components of this structure have been at least partially
qualified: landing mats for aircraft have been made of honeycomb,
and butyl rubber (in canal and reservoir linings) has demonstrated
weather resistance and toughness.
128
-------
O I O ! O i O j O i O j O ! O ! O i
A A A A A A A A A A
r' Y Y V y' V V V V V \
iojoioiOiOioiojoioloi
i o i o i o : o i o j o i o o o
-RUBBER SHEET
-HONEYCOMB
GRAVEL
Figure 17.
Hexagonal Honeycomb Covered with Perforated Rubber Sheet
Laid Over Gravel
RUBBER SHEET
HONEYCOMB
RUBBER SHEET
Figure 18. Covered Hexagonal Honeycomb with Second Perforated
Rubber Sheet Substituted for Gravel
129
-------
-------
SECTION XIX
PLANS FOR DEMONSTRATION PAVEMENT
SITE SELECTION FOR THE MOST PROMISING DESIGNS
(Phase B of Proposal)
Criteria for site selection will be developed. Some of them prob-
ably will be as follows:
(1) A local or state government must be willing to parti-
cipate on acceptable terms.
(2) The site should be representative of many in terms of
traffic and use.
(3) Weather information should be available in some detail.
(4) Soil and drainage information should be available or
readily obtained.
(5) The site should be large enough to install the test
pavement system and also a conventional system for
comparison.
(6) The site should be accessible, to permit inspection by
the public and its officials.
(7) If possible, there should be a special problem or advan-
tage heightening the value of porous pavement at the site.
(8) Administrative and/or security provisions should be feas-
ible for the conduct of the study and the protection of
any installations of equipment or instruments.
(9) A nearby
the pavement
fabricator, if required, would agree to produce
lent.
(10) Sites should be selected in various parts of the U.S.,
representing various climates, soils and pollutions.
It is expected that in the course of the feasibility study, the
attitudes and positions of public officials will be explored, and
from some of these we would expect to receive offers of cooperation.
It should not be difficult to find several candidate municipalities,
and to reach an agreement in principle on the EPA terms, for one'or
more test sites apiece. We would then visit each site to ascertain
that it presents no untoward difficulties of topography, size, drain-
age, soil information would be checked. These findings would be pre-
sented to the EPA for approval.
131
-------
Next step would be to take soil cores for examinations, to design
the pavements for the specific application, and to provide engineer-
ing specifications, complete with estimated costs. The municipality
would then be in a position to negotiate a grant with EPA, and to
arrange for the construction. We will consult as necessary during
these processes.
Included in the specifications would be provisions for instrumenta-
tion as described under Phase C. The size of the site is determined
in part by the monitoring and test requirements. A preliminary esti-
mate for a parking lot or shopping center is that the porous pavement
could occupy a four-acre square with one drain in the middle, by
which run-off, if any, can be measured. Beside this would be another
four-acre square with a drain in the center, covered with conventional
pavement, for a control.
EVALUATION OF DEMONSTRATION PAVEMENTS
(Phase C of Proposal)
The test pavements and their controls should be monitored for a full
year, to determine the effects of all seasons. Monitoring steps are:
(1) At each site there will be automatic monitoring of
weather (temperature, wind, and precipitation) to-
gether with runoff from the test pavement and the
control pavement and water level in the porous pave-
ment (see below). The difference in run-offs should
be equal to the amount imbibed by the porous pavement.
(2) Each site will be visually inspected twice a month to
ascertain any damage, cleanliness, evidences of change,
etc. During several storms, an observer will note how
the traffic proceeds, and how the pavement operates.
(3) At two locations in each test pavement and control pave-
ment, strain gauges just under the surface of the pavement
will measure its deflection, and FILPIPS (stress sensor)
on the soil under the gravel course will measure load.
A standard vehicle will pass over these instrumented
sections on occasion to measure load distribution and
deflection as affected by wetness of the soil, temperature
and age. On accasion, vehicles of different weights will
be used.
(4) At two locations near the center of each test pavement,
casings several feet deep (2 different depths) will per-
mit taking samples of water that has percolated through
the pavement and soil. These will be checked for pH,
coliform bacteria, oils, lead and solids. If possible
similar samples will be taken from similar installations
132
-------
in a nearby large lawn or field, and run-off samples
from the pavements also will be checked.
(5) At two other locations in each porous pavement, a per-
forated pipe will pass through to the soil. In this
pipe the level of water in the reservoir layer can be
gauged and samples of soil taken for bacterial cultures.
(6) Experiments will be run with sweepers and vacuum cleaners
to ascertain their efficiency in picking up trash and any
effect they may have on the pavements.
(7) Controlled contamination may be washed through the pave-
ment deliberately to see how it affects the purity of
the percolated water. At the end of the year, sections
of pavement can be removed and examined for contamination.
(8) Safety aspects of the wet and dry pavement, such as
breaking distances, skidding on curves, etc. will be
ascertained. Photographic evidence may be obtained, to
be shown to safety officials and others.
(9) Drivers, passengers, pedestrians and local residents
will be interviewed to determine public reactions to
the new pavements.
On each site will be a monitoring cabinet and weather instruments as
required in item (1). Items (3), (4), and (5) require installations
in the pavement where recorders will be connected during testing or
where grab samples will be taken. Items (2), (6), (7), and (8) rep-
resent ad hoc activities to be carried out on the site. Figure 19
illustrates the sampling arrangements.
The final report on each demonstration pavement will show, by means
of data and observation:
(1) Its adequacy to imbibe water and avoid run-off.
(2) Its ability to survive weather, loads, traffic,
and other conditions, and how its performance
changes with time and season.
(3) Quality of imbibed water, before and after per-
colation into the soil.
(4) Cleanability and maintainability.
(5) Effects of pollution.
(6) Judgements of comfort, esthetics, quietness, safety, etc.
133
-------
PRECIP WIND TEMP HUMID
WELL i
II 11 . II 1
s":/*:-**:!-
POROUS PAVEMENT
L^ 440 FT H
RUNOFF
CONTINUOUS
RECORDING
FLOW
1
1
RUNOFF ,
CONTROL PAVEMENT
L* 440 FT ^J
i ii ii -ir-.r-lTT
"v/*l a !«:"'°:'-"-V»»o"--'*
.i«.*j i-;«"»**vv"*i°{-*'
GRAB SAMPLING - NO INSTRUMENTS
SAMPLING FROM
a) STORAGE (SAMPLE PAVEMENT) 2 PER SITE
b) PERCOLATION (SAMPLE PAVEMENT) Z PER SITE
c) RUNOFF (CONTROL a SAMPLE) Z PER SITE
SOIL BACTERIA FROM
a) STORAGE (SAMPLE PAVEMENT)
VEHICLE WHEEL
STRAIN GAGE
GAGES INSTALLED (PORTABLE READOUT EQUIPMENT)
MECHANICAL TESTING (STRESS, STRAIN, RECOVERY,FATIQUE)
4 SITES (2 TEST PAVEMENT, 2 CONTROL)
Figure 19. Test at Site
134
-------
SECTION XX
ACKNOWLEDGEMENTS
This project was carried out within The Franklin Institute
Research Laboratories, by the following personnel:
The economics study is by Dr. Wilford C. Grover, Principal
Economist.
The Asphalt Concrete study is by Thomas I. Haigh, Project
Leader, and Dr. Arnold J. Hoiberg, bituminous consultant,
assisted by George Hawandjian.
The Portland Cement Concrete Work and the bacteriological
study are by R. H. Hollinger assisted by Peter Lydzinski.
Francis J. Sweeney worked on playground paving and on the
urban runoff study.
The project was conceived and supervised for The Institute
by Edmund Thelen.
Richard Field, Chief, Storm and Combined Sewers Branch, assisted
by Irving Seidenburg, Project Officer and Chief of Microbiology,
at Edison, New Jersey, monitored the work for the Environmental
Protection Agency under the direction and leadership of Frank
Condon, Project Manager and Staff Engineer. Muncipal Pollution
Control Technology Branch, EPA, Washington, D.C.
Mr. William A. Rosenkrantz, Chief, Municipal Technology Section,
E.P.A., provided many useful insights.
The support of this work by the Environmental Protection Agency
and the interest and contributions of these men, are acknowledged
with appreciation.
135
-------
-------
SECTION XXI
REFERENCES
1. (California Spec.) Standard Specifications, State of California,
Highway Transportation Agency, Dept. of Public Works, Division
of Highways, July 1964, p. 170, para. 39 - 206A Grading.
2. The Asphalt Institute, Manual Series No. 12 (MS-12), Asphalt in
Hydraulic Structures, March 1965, p. 18, Table II-2, Column 1.
3. Private Correspondence to E. Thelen from H. G. Spurr, Ministry
of Aviation Supply, Eng. Physics Dept., Royal Aircraft Establish-
ment, Farnborough Haunts, England, Dated April 2, 1971.
4. Clark, Proc. AAPT _27, 207, (1958).
5. "R.A.F. Aircraft Tests on Grooved, Open Graded, and Asphalt Run-
ways in Great Britain," B. Shilling, in Pavement Grooving and
Traction Studies, NASA SP-5073, 1968, pp. 67-75.
6. Private Correspondence: Vaughn Marker of the Asphalt Institute
to A. J. Hoiberg, March 8, 1971.
7. Federal Walter Pollution Control Administration Research Project
No. 120, by the American Public Works Assoc., Jan. 1969.
8. R. S. Means Co., Inc., 1970 Building Construction Cost Data, 1970;
Craftsman Book Co., The National Construction Estimator, 18th edi-
tion, 1970, McGraw-Hill Information Systems Corp., Dodge Con-
struction Pricing and Scheduling Manual, 1970. These secondary
sources have been used generally through Sections XIV and XVII.
9. Winfrey, Robley and Howell, .Phebe D., "Highway PavementsTheir
Service Lives," Highway Research Board Proceedings, 1967.
10. Stanford Research Institute, "The Economics of Asphalt for High-
way Construction," 1961.
11. "Supplementary Report of the Highway Cost Allocation Study," 1965,
89th Congress, First Session, House Document No. 124.
12. Moyer, R. A., and Lampe, J. E., "Study of Annual Costs of Flexible
and Rigid Pavements for State Highways in California," Highway Re-
search Record, No. 77, 1963.
13. Winfrey, Robley, "Highway Economics," Section 3, Highway Engineering
Handbook, Kenneth B. Woods, ed., McGraw-Hill, New York, 1960.
137
-------
14. Roy Jorgensen and Associates, "Virginia Maintenance Study>"
1966.
15. Ohio Department of Highways, "Ohio Maintenance Cost Study,"
1966.
16. Pennsylvania Department of Highways, unpublished maintenance
cost data.
17. Oglesby, C. H., and Altenhofer, M. J., "Economics of Design
Standards for Low-Volume Rural Roads," National Cooperative
Highway Research Report No. 63, 1969.
18. Camp, Thomas, R., "The Problem of Separation in Planning
Sewer Systems," Journal of the Water Pollution Control Federa-
tion, December 1960, p. 1961.
19. American Public Works Association, Problems of Combined Sewer
Facilities and Overflows, U. S. Department of the Interior,
FWPCA, 1967.
20. Ibid., p. 146.
21. Ibid., p. 75.
22. Ibid., p. 86.
23. Ibid., p. xii.
24. James F. MacLaren Associates, Report on Sewer Improvements in
the City of Toronto, 1960.
25. Benzie, W. J., and Conchaine, R. J., "Discharges from Separate
Storm Sewers and Combined Sewers," Journal of the Water Pollu-
tion Control Federation, March 1966, pp. 410-421.
26. Delaware Valley Regional Planning Commission, "Delaware Valley
Land Use Projections to 1985," unpublished material.
27. U. S. Water Resources Council, The Nation's Water Resources,
Government Printing Office, Washington, D. C., 1968, pp. 1-29.
28, Ibid., Section 6, pp. 1-20.
29. U. S. Army Corps of Engineers, "Middle Atlantic Water Resources
Plan," unpublished material.
30. Institute of Traffic Engineers, Parking Facilities for Industrial
Plants, Washington, D. C., 1969.
31. Martin, Norene, Executive Vice-President, National Parking Associa-
tion, personal communication.
138
-------
32. U. S. Department of Commerce, Bureau of the Census, Construction
Reports; Characteristics of New One-Family Homes, 1969.
33; Sumichrast, Michael and Frankel, Sara, A., Profile of the Builder
and His Industry, National Association of Home Builders, Washington,
D.C. 1970, pp. 100-112.
34. U. S. Department of Commerce, Construction Review, December 1970.
35. U. S. Department of Transportation, Highway Statistics, 1969.
36. School District of Philadelphia, School Site Standard Study, 1969.
37. U. S. Department of Commerce, Bureau of the Census, 1970 Statis-
tical Abstract of the United States.
139
-------
-------
SECTION XXII
GLOSSARY OF TERMS RELATING TO PAVEMENT
Aggregate - rocks, usually crushed, of specified size distribution
and durability properties.
Aggregate, coarse - retained on #8 sieve.
Aggregate, fine - passes thru #8 sieve.
Asphalt, asphalt cement - a black, viscoelastic, waterproof residue
from the distillation of petroleum. It also occurs naturally in
some parts of the world.
Asphalt Concrete - A paving material consisting of aggregate bound
with asphalt, made by heating both materials to around 300°F, followed
by mixing, delivering, spreading, and compacting while still hot.
Bituminous - a class of materials of a pitch-like nature, including
asphalts, coal tars, pitches, etc.
California Bearing Ratio (GBR) Test - Quantifies the load-bearing
capacity of a soil on a scale ranging from zero (no capacity) to
100, which is the load-bearing capacity of crushed stone.
Cutback - asphalt dissolved in a petroleum distillate, which eva-
porates after laying, as the asphalt "sets".
Dense graded - Usual aggregate distribution, with enough fine
particles to reduce the void volume to below 4% in an asphalt concrete,
Durability test - As used in this report, a test to determine the
extent to which the asphalt in a porous pavement hardens when air
at 140 or 150°F is passed thru the pavement for several weeks.
Emulsion - A suspension of asphalt particles, generally 1 to 2 mic-
rons in diameter, in water. Generally contains 55-62% asphalt by
weight.
Freeze-thaw test - A test in which the pavement is filled with
water, drained, cooled to 0°F and the heated to 50°F. This cycle
is repeated enough times to simulate the temperature cycling
naturally occurring during years of service.
Gradation - The distribution of particle sizes in aggregate.
Marshall test - ASTM Test D-1559. A test for measuring the load
required to crush a standard cylinder of pavement, and the degree
of deformation, or flow, at which the cylinder ruptures.
141
-------
Open graded - Aggregate containing relatively small amounts of fine
particles. Generally contains over 10% by volume of voids in an
asphalt concrete.
Pavement - A durable, tough layer of material subjected to vehicular
or pedestrian traffic; Normally it is relatively waterproofed.
Permeability Coefficient - The volume of water, in cubic feet, under
a head of one foot, that will pass thru a square foot of porous sur-
face in 1 day.
Penetration test - ASTM Test D-5. A means of determining the hard-
ness of bituminous materials by measuring the distance a standard
needle will penetrate into them in a given time, under standard
load and temperature. Usually 5 seconds, 100 grams, 77°F. Asphalt
grade 85 to 100 pen. means the needle will penetrate from 0.85 to
1.00 centimeters.
Porous pavement - A pavement through which water can flow at signifi-
cant rates.
Seepage rate - Calculated from the permeability coefficient, is the
inches of rain that would pass through the pavement in one hour, with
no water standing on the surface. A head exists due to the thickness
of the pavement.
Stripping - The tendency of water to displace asphalt from stone.
Stones vary in their stripping resistance and anti-stripping agents
are added to the asphalt if the stone is deficient in this respect.
Subbase - A bed of aggregate under the pavement, which helps to dis-
tribute wheel loads over the underlying subgrade. This aggregate
sometimes is lightly bonded with asphalt.
Subgrade - The soil under the subbase. If the soil naturally occurring
at the site is deficient in load bearing capacity it may be treated
or replaced.
142
-------
1
Accession Number
w
5
2
Subject Field & Group
Selected
Resources
Water
Abstract
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
s
Urbanization
The Franklin Institute Research Laboratories, Philadelphia, Pa.
Colloids and Polymers Laboratory
Title
Investigation of Porous Pavements For Urban Runoff Control
10
42
Authors)
DT W"l "\ "FnTrl P nvrftTov
Dr. Arnold J. Hoiberg
Mr. Thomas I. Haigh
Mr. Francis J. Sweeney
Mr. Edmund Thelen
Citation
1 £ Profe-ct^De^tir^tjon
1103^- DOT
2] /Vote
23
Descriptors (Starred First)
Combined Sewers; Treatment Plants; Storm Sewers; Pavement; Water Conservation;
Urban Runoff; Solid Waste Disposal; Asphalt Concrete; Traffic Safety; Community
Planning.
25
Identifiers (Starred First)
Porous Pavements
27
Abstract
Laboratory studies demonstrated the technological and economic feasibility of
open-graded asphalt concrete as porous pavements to conserve water, to reduce
loads on combined sewers and treatment plants, and to mitigate urban runoff as
a destructive environmental factor.
The disposal of demolition wastes in these pavements is a plus factor.
'Abstractor
h Kdmund Theleii
Institution
The. Franklin
Institute
SEND, WITH COPY OF DOCUMENT, TO: WATER RESOljRCTES^t;fgWrWli! INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
ft U.S. GOVERNMENT PRINTING OFFICE. 1978 757-140/1316
-------
-------
-------
g-S =5
o ^ -c
^ 9 Q
co S. c:
Qo o
r, 5 ^
* . O Cl) Q)
""* ^ < S.
O <*> CD CD
*** 5 o> S
Is. |"
*IPI
§
u
^
--). "^
I
CD
s.
CD
Z T3
m
H 3 CD
c < c
z > co
3 m z
< .- rn
13 -
O co
< °
m °
CD
O
O
3T
CO
TJ
CD
O
Tl
O
o
(A
0)
CO
51 O rn
Q)
CD
j
>
O -i Q. O
' +1 ^^ I
ft § o m
I H§
-3 O __
o
Cl)
3
m
>
O
m
z
O
c
cn
rn
O O
7 (/)
s £
"7 CD
m -H m
TJ > >
f r z
co -a o
CO 11
01 <3 S
-H m
rn (fi
*3 T
2 >
i 5
Q
m
2
O
S?
------- |