|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Environmental
NRMCA Pervious Concrete Contractor Certification
|
Hydrologic Design of Pervious Concrete Part 1 w Part 2 w Part 3 w Part 4 w Part 5
Hydrological Design Methods
4.1 Introduction
The methodology used in the hydrological design of pervious concrete pavement systems should reflect the level of detail needed to satisfy the agency specifying, permitting, or regulating the use of pervious pavement. The methodology should also be sufficiently rigorous to meet the needs of the design professional, and should reflect the behavior of the system in service within the limits of accuracy needed. Computational efficiency is a desirable, although not compelling, factor, and model complexity may not necessarily improve model accuracy. A model that is simultaneously simple to use and captures the essential elements of behavior is useful and important to the design professional, even when advanced analysis is not required by local regulations.
Site conditions and regional needs can vary significantly. Local regulations can range from simple to complex depending on the needs and characteristics of the area and the objectives of the regulatory agencies. Solutions and approaches suitable for one area may be overly restrictive and prescriptive in another, or provide insufficient protection in a third.
The method recommended in this document is the NRCS (SCS) Method, or “Curve Number” method as outlined in Technical Release 55 (TR-55) (SCS 1986). This method:
w Is well established and widely used by many design professionals involved in managing runoff,
w Captures the essential elements of pervious concrete pavement system behavior,
w Is appropriate for the design of a structure intended to capture and hold some portion of the runoff in a small urban watershed (such as a retention or detention feature),
w Is flexible and easily adapted to a site with several types of surfaces contributing to runoff,
w Is easily implemented by adapting well known stagestorage-discharge principles to the simple geometry of a pervious pavement system, and
w Can be used to analyze systems intended to function within the constraints of many different regulatory requirements.
Users of this document who are unfamiliar with hydrological design methods should be aware that there is no nationally accepted, standard design technique for estimating total runoff; preferred techniques vary with region and application. Techniques favored in the western US are generally those of the Bureau of Reclamation, while those favored in the eastern two thirds of the US are often those of the Natural Resources Conservation Service (NRCS). Results with these methods are similar enough that the techniques presented in this document based on the original NRCS (SCS 1986) methods can be adapted for most applications. Other potential hydrological design methods not reviewed in this document include the Chicago (Tholin's) Hydrograph Method (Tholin and Keifer 1960), the Illinois Urban Drainage Area Simulator (Terstriep and Stall 1974), the U.S. Army Corps of Engineers' Storage, Treatment, Overflow Runoff Model (STORM) (USACE 1977), and the Storm Water Management Model (SWMM) (Rossman 2005), many of which estimate peak flows.
Another common approach, the Rational method, is also discussed for completeness. In the authors' opinions, the Rational method, while acceptable and appropriate in many regions or situations, is not the best methodology to use when analyzing pervious concrete pavement systems. The results must be used with caution and can lead to problems in some situations if used without considering all aspects of the system behavior. These limitations are discussed briefly in Part 5.
Since the design of pervious concrete pavement systems typically involves hydrological design of relatively small watersheds for very specific purposes, and since many of the input data are known or estimated with limited precision, the use of a complex model provides neither additional accuracy nor additional information or insight into the solution, while the computational cost can increase significantly. A relatively simple but flexible model is adequate and appropriate for these applications. The methods described below, especially those in Section 4.4, are suggested since they are well established, easily implemented and are commonly employed in many parts of North America.
The authors have attempted to keep this review relatively simple and still provide sufficient fundamental technical background and discussion to assist in developing a useful pavement structure. Other methods are not reviewed in detail in this document since they are not commonly used, are limited in applicability, or are overly complex and intended for analysis of much larger watersheds (Corps of Engineers methods, for example fall into this later category). Clearly, the design professional experienced with these or other design methods can provide a structure with satisfactory performance. This document is not meant to substitute for the experience and professional judgment needed to fully design a complete system, including both a pervious concrete pavement and overflow structures. It is intended to be used to craft those designs more efficiently and help produce structures that instead simultaneously meet the needs of the site owner, regulatory agencies, and the community.
4.2 Percent Impervious Surface Since the percent of impervious surface in a watershed directly affects the quality of streams in that area and downstream (see Figure 13), a simple, easy to implement land management policy is to limit the amount of impervious surface in the built up area to some specific limit. In some jurisdictions, therefore, regulatory restrictions may limit only the percent of impervious surface in a watershed. In these situations, advanced analysis may not be required; pervious concrete pavement systems should be considered to be pervious areas in determining total impervious area. In other jurisdictions, additional constraints and limitations are placed on allowances of pervious pavement systems in calculating the percent impervious area. In these and many other situations, a more detailed hydrological analysis of the project is required.
4.3 The NRCS Curve Number Method
The Curve Number method is often combined with stagestorage-discharge methods to design impoundment features such as retention and detention ponds. While stage-storagedischarge functions can be complex and time consuming to formulate in general, the simple geometry and discharge characteristics of flat pervious concrete pavement systems make the TR-55 (SCS 1986) Tabular Hydrograph Method easy to adapt to spreadsheet analysis, and easy to use in practice. Therefore the Curve Number method can provide a realistic, robust, and easily implemented model of the hydrologic characteristics of a site while incorporating the effects of pervious, impervious and other surfaces with a variety of cover, to estimate the total runoff and the total volume of rainfall captured and infiltrated.
4.3.1 Curve Number Method — Design Methodology
The Curve Number method estimates the total volume of runoff, Q* (inches) using Equations 4 and 5 below. The runoff volume is designated “Q” in the literature, but the “Q*” designation is adopted in this document to distinguish it from the peak flow (ft3/s) Q estimated by the Rational method.
4.3.2 Curve Number Method — Design Input
The volume of runoff can be estimated using the assumptions and methods described in TR-55 (SCS 1986) with two sets of data– precipitation volume and CN of the area or subareas. The distribution of rainfall in the storm is based on general geographical location.
4.3.2.1 Curve Number Method — Design Storm
The value for P is the total volume of precipitation expected in the design storm. Traditionally, the 10-year storm has been used in the design of stormwater collection systems (Veissman and Lewis 2003), with the 2-year storm often considered the “service load” storm for the site. The pervious concrete pavement system designed for active mitigation must be integrated into a system designed for the 10-year storm, including overflow structures. For new construction, this means that pipe sizes required for a 10-year storm can often be reduced with active mitigation using pervious concrete pavement systems, resulting in a cost savings. It may not be necessary to increase the capacity of existing storm sewers with additional development or retrofit applications when an active mitigation system using pervious concrete pavement is utilized.
As a general guideline, the storage capacity of an active pervious concrete pavement system is designed to accommodate most, if not all, of the site runoff of the 2-year, 24-h rainfall. The performance of the system is then checked in the 10-year, 24-h rainfall, as a minimum. Some jurisdictions require that performance in other storms must be checked as well.
The total volume of rain is clearly important; however, the effects of infiltration into the soil over time must also be considered and, therefore, the distribution of the rainfall over the 24-h period (the hydrograph) must be included in the design. The precise shape of the hydrograph is not critical, and the use of the NRCS design rainfall event is suggested unless the design professional believes another method would be more appropriate. Hourly increments are appropriate for pervious concrete pavement system design.
The NRCS design storm is a center weighted, 24-h, unit rainfall event, with various rainfall intensities per hour appropriate for various regions with different types of storms. Types I and IA are consistent with rainfall patterns in the Pacific maritime climates with wet winters and dry summers; Type IA gives the least intense rainfall of all types. Type III should be used in the coastal areas of the Atlantic and Gulf of Mexico where tropical storms with large 24-h rainstorms occur. Type II storms are appropriate for most of the United States (See Figure 15). Type II storms also have the most intense, short duration rainfall segments and so can be used conservatively for Type III areas as well. See TR-55 (SCS 1986) for more information on these classifications. An important advantage of the NRCS distribution is that segments forming the 24-h pattern also comprise the design 1-h, 2-h, and 6-h storms so that the performance of the system under design can be evaluated at all of the intervals of interest by using the NRCS design storm (See Figure 14 and Malcom 2002).
The unit ordinates of the hydrograph for each time period are multiplied by the appropriate storm depth for the location of interest. This produces a design rainfall event over time that provides the total volume and distribution of intensities appropriate for that particular location. For example, if the total volume of rain in the 2-year, 24-hour storm for a location in the Type II area was 3.6 in. (91 mm), the anticipated rainfall in the first hour would be only 0.04 in. (1 mm), while the total rainfall in the 12th hour (the middle of the storm) would be 1.54 in. (39 mm), which is also the maximum hourly rate of precipitation for this storm type.
4.3.2.2 Curve Number Method — Definition and Values
The NRCS Curve Numbers (SCS 1986) are used to estimate the runoff of an area or sub-area with a given type of cover, over a given soil, for a given depth of precipitation. A higher CN means more runoff: a CN of 100 means that all rain will runoff. CN’s are no greater than 98, even for conventional pavements, since some small amount of rainfall will be held by the surface. By using coefficients (CNs) based on both soil and cover characteristics, the Curve Number method provides a more flexible and site specific method of selecting appropriate design values for estimating runoff than the use of Rational method coefficients.
The NRCS provides tables to estimate the CN of various areas with a given type of cover for soils classified, for hydrologic purposes, as Hydrologic Soil Group (HSG) A (sand, loamy sand, or sandy loam), HSG B (silt loam or loam), HSG C (sandy clay loam), or HSG D (clay loam, silty clay loam, sandy clay, silty clay, or clay), as described in Chapter 3. The soil horizon with the highest infiltration rate is HSG A; the infiltration rate is lowest for HSG D. Various charts are also available, such as that shown in Figure 6, to assist the designer in selection of an appropriate CN.
The CN for open space in good condition (more than 75% grass) in developed urban areas ranges from 39 to 80, depending on the soil type. Woods and grass cover, such as found in orchards in an agricultural area, generally have CN’s which range from 32 for cover in good condition over sandy soils with excellent drainage capacity to 86 for cover in poor condition over poorly draining soils (see Appendix A). It is clear that the characteristics of the underlying soil play an important role in the expected runoff in any particular site, with or without pervious concrete. These same soils form the subgrade under a pervious concrete pavement and, therefore, affect the rate at which rainfall captured by the pervious concrete infiltrates into the soil.
4.3.3 Curve Number Method — Design Procedure
The design procedure used with the Curve Number method is also described in CD063 (PCA 2007). In addition to design and analytical tools such as those described below, CD063 also contains typical values for the 2-year and 10-year, 24-h storms for a variety of locations, plus excerpts of documents such as TR-55 to assist in selection of an appropriate CN.
In general, the Curve Number method consists of mathematically applying the hourly distribution of rainfall for the design storm to the various surfaces of the site that discharge onto the pervious concrete pavement system. For an active mitigation system, this can include impervious surfaces such as building footprint, paved islands, and bus or truck lanes, and surfaces with natural cover such as planted traffic islands, vegetated areas on site, and adjacent properties that drain naturally onto the site under design. For a passive mitigation system, this would typically include only the surface of the pervious concrete pavement, but may also include border features associated with the pavements, such as curbs or impervious decorative borders.
The volume of rain for each hourly increment of the design storm falling on the pervious concrete and the impervious surfaces, and the excess surface runoff from adjacent areas (Q*, based on the CN of the contributing area) less the volume infiltrated into the soil, is stored (impounded) in the pervious concrete pavement system. Overland flow occurs very rapidly for small sites, so no adjustment is made for travel times for contiguous areas. This is both computationally convenient and conservative.
This process continues until the rainfall of all of the 24 hourly increments has been applied or until the storage capacity of the system has been exceeded, in which case the remaining rainfall is considered to be excess surface runoff. The procedure can be easily implemented on a spreadsheet (see Appendix B).
Infiltration maintains the effective storage capacity of the pervious concrete pavement system by removing some of the rainfall over time. The effect of infiltration on storage capacity, and therefore excess surface runoff, is a critical element in the analysis. Infiltration continues until the pervious concrete system is emptied and the storage capacity returned to its original value. The total recovery or drawdown time (the time until 100% of the storage capacity has been recovered) is also an important performance factor.
The system must be emptied and full storage capacity recovered in a reasonable amount of time. This is often the limiting factor for active mitigation applications in poorly draining soils. Recovery time is typically not a major concern in passive mitigation applications with these types of soils since the infiltration is slow and runoff relatively high even with natural cover. A recovery time of 5 days or less is reasonable for active mitigation, considering the limited probability that another significant storm will occur within 5 days. This also is common practice in water quality engineering.
In situations where recovery time is excessive, and the pervious concrete pavement system is intended to handle runoff from surfaces in addition to the pervious concrete pavement, the storm sewer system must be designed to carry essentially all of the runoff. In these cases, the pervious concrete pavement system may still provide a useful hydrological function by capturing much of the first flush.
The design professional should analyze a range of infiltration rates and design storms, conducting sensitivity analysis of performance under a variety of conditions. If the preliminary design indicates borderline performance, additional on-site investigations of infiltration may be useful. Percolation tests may not provide the needed information if conducted on the natural soil, however, and so may provide only marginal benefit. Percolation or other tests conducted on the compacted soil to be used on site should be used to confirm the general accuracy of the initial estimates when the design indicates marginal performance. As a preliminary indicator, if the site is geologically suitable for a septic system, it is probably suitable for pervious concrete pavement systems in active mitigation applications, although this guideline may be overly restrictive as a policy statement. Once verified on site, and recognizing the limits of accuracy in the design assumptions and model, a prudent designer should modify the design such that it is no longer “borderline” rather than try to improve the accuracy or precision of estimate of infiltration with additional extensive testing.
4.3.4 Curve Number Method — Output
Application of the Curve Number method to sites with pervious concrete pavement systems should provide at least two results – the total runoff from the site (in inches or mm) and the system recovery time (in days). The runoff can be converted to acre-feet or cubic feet (or cubic meters) knowing the area of the site. The excess runoff can also be converted into an equivalent CN if desired. The hourly runoff in the design storm may also be useful in some analysis.
The “equivalent CN” can be calculated for a given site based on the precipitation and the estimated volume of runoff from the site using Equations 4 and 5. For example, if 3.6 in. (91 mm) of rain fell in the 2-year, 24-h rainfall, and the pervious concrete pavement system held all but 1 in. (25 mm), which became excess runoff, the equivalent CN would be approximately 69. This value should be used with caution since the CNs developed by the NRCS were functions of the soil and cover characteristics and an “equivalent CN” calculated as just described must be a function of volume of precipitation. Since different storms will result in different values, the equivalent CN is best used for comparisons of various alternatives or to compare against the pre-development CN of the site with the same storm. The equivalent CN may also be used with additional TR-55 based analysis of downstream sites or elements, however, as discussed in 5.5.
Excess runoff is anticipated for soils with natural cover in any practical storm of interest. By impounding a significant portion of the runoff from a site, the hydrologic characteristics of a site containing a pervious concrete pavement system can, when designed to meet that specific goal, resemble those of the same site prior to development. For example, an active, grassy pasturage used for grazing livestock, in fair to good condition hydrologically, on silty soil, with a moderate infiltration rate, might have a CN of 66 prior to development. It can be shown that this same site after significant commercial development (about 30% parking area composed of pervious concrete and base course, about 15% vegetated islands, and about 55% impervious pavement and roof structure), could maintain a similar equivalent CN with a properly designed pervious concrete paving system.
4.4 The Rational Method
The “Rational Method” is commonly used to estimate the maximum runoff rate that will occur at any one time and place in small urban watersheds (those less than about 1 mi2 (about 2.5 km2)) (See Veissman and Lewis 2003, for example). This method is a simple technique, long used for estimating the maximum or peak flow (volume per unit time) anticipated from a storm that must be handled by culverts, swales, storm sewers and other “outlet” or drainage features. The Rational method estimates the maximum flow expected at some location (the outlet) rather than the total amount of runoff, and so must be used with caution in assessing the performance of retention or detention structures such as pervious concrete pavement systems. It is useful to briefly review the Rational method in more detail because it is the basis of many designs for storm drainage facilities and may be selected as the design method of choice for pervious concrete systems in some situations.
When used to analyze watershed behavior in which flat pervious concrete pavement systems overlay sands with moderate to high infiltration rates, the Rational method will often provide acceptable results. However, the Rational method may not fully capture all of the advantages of pervious concrete paving systems and can lead to problems in implementation and interpretation when used in complex situations. The Rational method should therefore be used with caution when applied to pervious concrete pavement systems. Additional discussion is provided in Section 5.8.
4.4.1 Rational Method — Design Methodology
In the Rational method, peak flow is estimated using the relationship:
A higher value of C means more of the rainfall is expected to runoff the surface being analyzed; a value of “1.0” would indicate that 100% of any rain falling on that surface would run off the surface, for example. Conventional pavements are typically assigned a C of 0.98, indicating that almost all of the rain falling on that pavement would become runoff (some small amount is captured in wetting the surface and held in depression storage such as the irregularities and small “birdbaths” found on most pavements). The values of C of the individual areas are empirically based and effectively non dimensional when using customary US units. The value of C for an individual area is normally derived from tabular data, and adjusted by experience in a given location, and varies within limits, by application.
The value of C used in the equation is a composite or average value, of Cs of smaller, individual areas, weighted by area. For example, if:
the C (composite) used in the equation would be:
4.4.2 Rational Method — Design Input and Use
Since small urban watersheds are sensitive to short, intense rainfalls, the design storm selected is one of relatively short duration when using the Rational method. The duration of the design storm is equal to the “time of concentration” of the watershed. This is the amount of time necessary for a drop of rain to flow from the farthest point in the watershed to the outlet structure being designed. The basis for this approach is that a storm lasting the time of concentration will be the shortest (and therefore the most intense) storm which will still fully contribute to runoff at the outlet structure.
Various methods may be used to estimate the time of concentration. Kirpich's equation (Kirpich 1940) is simple and, although somewhat dated, still provides estimates with sufficient reliability to be used for designs in small watersheds. TR-55 also provides a method for estimating time of concentration, although this estimate tends to be lower (faster flow) than many others. The time of concentration for most small, urban watersheds is 15 to 30 minutes, and rarely as much as an hour.
Different, experienced designers, working independently, will almost inevitably arrive at slightly different values of peak flow. This is rarely a problem in the design of outlet structures such as culverts commonly employed in small urban areas because the pipe used in these structures are available in only a few select sizes. The selection of an appropriate value of C for a pervious concrete paving system can also be expected to vary somewhat.
Adjustments should be made in the value of C for storms with different return periods, although the value of C is the same for the 2-year and 10-year storms (Wright-McLaughlin 1969). For example, if C in the 2-year storm was 0.85 for a given site, the C used in analysis of the same site in the 50-year storm would be 1.0 (C can never exceed 1 .0). If the C in the 2-year storm was 0.50, the C used in analysis of the same site in the 50-year storm would be 0.6.
In some cases, there may be a distinction between “connected” and “disconnected” impervious areas when estimating the time of concentration. A disconnected impervious surface is one surrounded by natural ground cover and not connected directly to the drainage channel or storm sewer. The benefit of disconnected impervious areas is that the volume and velocity of discharge of the drainage channel is reduced compared to that expected when the impervious surface drains directly into the channel due to the buffer effect of natural ground cover. A much simpler analytical approach is to make no distinction between the two.
Table 3. Typical Adjustment Factors for Rational C for Storms at Selected Return Periods
Reference: Leming, M.L., Malcom, H.R., and Tennis, P.D., Hydrologic Design of Pervious Concrete, EB303, Portland Cement Association, Skokie, Illinois, and National Ready Mixed Concrete Association, Silver Spring, Maryland, USA, 2007, 72 pages. |
ConcreteAnswers for Architects, Engineers and Developers:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
About NRMCA | Privacy Statement © National Ready Mixed Concrete Association, 2010
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
.jpg){kind=small}
