 |
|
|
|
|
|
|
|
|
|
 |
 |
||||||||
 |
 |
 |
 |
||||||
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
 |
|
Pennsylvania Handbook of Best Management Practices for Developing Areas§5. BMP Design §5.1 Watershed Studies The employment of BMPs should be done within the context of an assessment of the specific needs of the watershed. Common effects of development at the watershed level include:
The severity of predicted or realized impacts will influence the development of criteria for controls in a watershed. To facilitate implementing a watershed management program, watershed studies should be prepared for an entire watershed. Typically, a watershed will encompass more than one county and numerous municipal jurisdictions. Therefore, the development of watershed-wide plans will involve cooperation and compromise on the part of participants. Watershed studies generally are done under the auspices of the Pennsylvania Act 167 stormwater management planning grants program. The recommendations of the watershed plans are implemented at the local level. The watershed studies should include preparing hydrologic and hydraulic models and analyzing water quality and wetland impacts. The recommendations from the watershed studies include, in most cases, structural and nonstructural BMPs that provide flood protection, meet stormwater quantity and quality criteria, prevent erosion and sedimentation problems, and preserve wetlands to the maximum extent possible. Outcomes of the studies include the development of ordinances governing the criteria for onsite controls in the watershed. The criteria, which frequently specify design storms for BMP sizing, are the basis for selecting and deploying BMPs in the watershed. The watershed studies also are useful for identifying monitoring sites, and assisting in preparing monitoring plans for water levels, discharges, toxics, nutrients, etc. Occasionally, a watershed study will identify conditions where onsite controls will not adequately protect the watershed. Examples include watersheds where:
In these instances, the results of watershed studies can be used for developing regional stormwater management projects that can help achieve goals and requirements for the entire watershed. Riparian corridor management is an effective BMP for controlling flooding on a region basis. Effective management of riparian corridors for providing wetlands, overbank floodwater attenuation storage, flood control weirs, or forested water quality buffers, requires accurate stormwater models of the watershed. Regional facilities are not appropriate methods for achieving groundwater recharge objectives. Some advantages of regional stormwater-management facilities are the following:
A significant difficulty in developing regional capital projects is funding, which requires the participation and cooperation of all watershed residents and municipal governments. Planning, legal, land acquisition and design costs may be significant. Extended delays in implementing the development of regional facilities resulting from these impediments may imperil the watershed as development continues. Efficient on-site design using BMPs generally will result from an orderly five-step process:
A common error is made in site planning when designers begin by considering runoff peak attenuation. This often leads to "end-of-pipe" measures that consume large areas, are expensive to maintain, and conflict with the aesthetic objectives of the architect. Although, site-specific conditions may vary, the following design principles will remain the same. §5.2.1 Break Up Large Impervious Areas Site designers should look for opportunities through natural topography, grading, or segmentation to break continuous impervious surfaces into more than one drainage area so that stormwater is delivered to several edges or points and is not concentrated at one discharge location. The larger an impervious surface is that slopes to the same edge or point, the larger is the stormwater management challenge. Therefore, parking lots, roads, and roofs, should be designed so that runoff is divided among several outlets or treated before it reaches these outlets. This creates a diversity of opportunities to apply different stormwater management measures that can be better fit into the landscape of the development site. §5.2.2 Apply BMPs near the Source of Runoff The placement of control measures, including devices for the detention, infiltration, and treatment of runoff (such as bioretention facilities), close to the runoff-generating surfaces returns water to its natural location in the landscape and supports the normal soil-water balance in immediately downstream areas. Furthermore, by intercepting and controlling runoff before it becomes concentrated, the need for downstream mitigation measures (e.g., energy dissipaters, lined channels, and storm sewers) can be minimized. The efficiency of all control measures is improved when the volume of water that must be accommodated is minimized. Opportunities for incorporating BMPs in the site layout should be identified early in the design process. Costs generally can be saved by careful planning the site to place control measures near to the runoff-generating impervious surfaces. Large concentrated flows ultimately require larger end-of-pipe facilities to meet quality, groundwater recharge, and peak-control goals. Large facilities are more difficult to site and do not easily fit into the landscape design. §5.2.3 Evaluate Needs for Treating Runoff (Per Local Ordinance) Identify impervious surfaces that will produce runoff with a potential to degrade water quality. In consultation with the municipality, determine the pollutants and select appropriate treatment measures. Establish specific performance criteria for designing the measures. Search for areas immediately adjacent to these surfaces where water quality inlets, bioretention areas, or other selected treatment measures can be conveniently located. Preliminarily size the measures by using the water quality design storm. Determine if enough area is available for installing appropriate water quality devices to treat the runoff derived from the water quality design storm. If required, consider redesigning or relocating runoff-generating surfaces to provide enough space for water quality measures. Evaluate the ancillary benefits of the selected water quality BMP, which may include capturing runoff or attenuating peak runoff. By taking these benefits into consideration, the size of other BMPs may be reduced. §5.2.4 Satisfy the Groundwater Recharge Objectives (Per Local Ordinance) Search for areas adjacent to each impervious surface where infiltration devices can be installed. Below-grade infiltration trenches or seepage beds are only one type of infiltration measure. The most direct application of the principle is to convert an impervious surface to a pervious one by using permeable-pavement systems in parking lots, walkways, and public pedestrian spaces. These measures deliver precipitation directly to the soil beneath the surface. If permeable-pavement systems are not feasible, sheet flow from impervious surfaces can be directed to the long edge of above- or below-ground infiltration devices such as bioretention facilities or gravel-filled trenches. If the runoff capture tables are used for design (see Appendix F, Runoff Capture Design), sum the volume of total runoff capture by the selected BMPs and deduct the total from the site runoff capture requirement. All devices that retain runoff for groundwater recharge should be capable of completely infiltrating the impounded water within 48 hours. Most of the measures also will improve water quality by the filtering action of soil, nutrient uptake by vegetation, and processing of pollutants by microorganisms. §5.2.5 Satisfy the Runoff Peak Attenuation Objectives (Per Local Ordinance) The design of runoff peak attenuation measures requires determining the predevelopment peak-runoff rate at the site boundary. The postdevelopment peak-runoff peak then is computed taking advantage of the attenuating properties of the proposed site design, including water quality and runoff-capture measures. Peak runoff can be reduced by using a variety of drainage modifications that are associated with other BMPs, including:
In most instances, the combined effect of the measures will substantially reduce the peak flow rates. If the postdevelopment peak-runoff rate at the site boundary determined in this manner is predicted to be higher than the predevelopment rate for the design storm, then runoff peak attenuation measures must be included in the plan. At most sites, some form of runoff detention facility will have to be constructed to satisfy the requirement for maintaining predevelopment peak-discharge rates of the larger design storms. However, the size and complexity of the devices generally can be reduced by detaining runoff close to the impervious surfaces. The design storm approach is one method of establishing uniform criteria for runoff management design. Throughout the handbook, four types of design storms are referred to. In order of increasing magnitude, the design storms are:
Runoff capture design storms are benchmark rainfall events used to develop criteria for groundwater recharge. To comply with the criteria, it must be shown that no appreciable runoff will occur during the runoff capture design storm. The volume of water that must be retained, the "runoff capture volume," can be expressed as inches of rainfall over the project area. BMPs that can be used to enhance groundwater recharge, and are appropriately evaluated by using the runoff capture design storm, include permeable pavement, bioretention, and infiltration trenches and dry wells. Water quality design storms are used to size BMPs that are intended to achieve specific quality treatment objectives. Examples of BMPs with significant water quality functions are:
Criteria based on water quality design storms generally require that the design treatment efficiency be achieved during the water quality design storm and all smaller events. For instance, the overland flow velocities used to evaluate the efficiency of filter strips for trapping sediment should be based on the runoff predicted for the water quality design storm. Although water quality BMPs usually will continue to provide benefits during runoff events larger than the water quality design storm, water quality BMPs are not required to achieve the same level of treatment for these events. Design for the attenuation of runoff peaks is traditionally based on the 2-, 5-, 10-, and 25-year return-frequency storms. These large storm events are capable of producing significant downstream flooding and streambank erosion. The design criteria requires that the predicted postdevelopment peak runoff rates for the runoff peak attenuation design storm does not exceed the peak associated with predeveloped conditions. A wide variety of BMPs can be used to control peaks. The BMPs include:
Runoff peak attenuation design storms are usually based on the 24-hour rainfall. Shorter rainfall distributions may not adequately account for partial filling of detention storage prior to the occurrence of the peak storm runoff. The selection of appropriate design storms is the prerogative of Pennsylvania municipal governments. Design storms usually are included as part of the stormwater ordinances developed by the townships. Ideally, design storms will be tailored to the local site conditions. The Pennsylvania Act 167 Storm Water Management Planning Grant program offers one mechanism for municipalities to develop stormwater management criteria that are based on local needs and objectives. Different approaches have been suggested for selecting design storms. In Appendix F, one of the approaches, which is based on an analysis of the cumulative volume of annual rainfall, is discussed. The Natural Resources (formerly Soil) Conservation Service’s (NRCS) 24-hour type II rainfall distribution for the specific locality is recommended as a basis for generating design storms. The use of a fixed storm duration provides for a consistent comparison of runoff volumes computed for various storm intensities and drainage subareas. Design storm hydrographs can be calculated by using the NRCS methodology described in the NRCS National Engineering Handbook, Section 4. This methodology includes the so-called soil cover complex and nondimensionlalized unit hydrograph techniques and is implemented in a variety of computer simulation packages. Alternative methodologies, including kinematic wave runoff routing and synthetic unit hydrograph generation, also are available in various computer software packages. Some frequently used computer programs for hydrograph analysis include TR-20, HEC-I, SWMM, and PSRM. §5.3.1 Runoff Capture Design Storm Impervious surfaces generate concentrated flows that go directly into streams and deny precipitation its natural route into the soil and water table aquifers. The consequence is that flow to wetlands and streams will be severely reduced during non-storm periods. Application of a runoff capture requirement is a practical way to achieve the goals of maintaining aquifer recharge and protecting stream base flows. The criteria recommended for runoff capture is that the percentage of annual rainfall that is infiltrated should not change significantly after development. For instance, a watershed may have an overall runoff annual ratio of 40 percent for predeveloped conditions (i.e., 60 percent of annual rainfall is retained and infiltrated). The associated runoff capture design storm will have the property that 60 percent of the annual rainfall will occur in storms of equal or smaller magnitude. Therefore, controlling the runoff from the water quality design storm is sufficient for preserving the overall water budget for the watershed. The hydrology of larger storms does not need to be evaluated. This approach requires that the runoff ratio that characterizes the undeveloped or natural watershed be estimated. The estimates can be done by constructing a water budget. Alternatively, an approximate runoff ratio that reasonably characterizes the watershed can be drawn from literature. The prevalence of low-permeable soils (i.e., NRCS hydrologic soil group D), steep slopes, or thin soil result in higher runoff ratios. In Pennsylvania, runoff ratios can be expected to range between 25 and 50 percent of total annual rainfall. A method for computing the magnitude of the runoff capture volume is presented in Appendix F. When a runoff ratio of 40 percent is evaluated for the five rainfall regions in Pennsylvania (Aron, et al., 1986), the runoff capture volume computed by using this method ranges between approximately 0.46 and 0.83 inches. Storms of this magnitude will occur frequently. Compliance with the runoff capture criteria can be demonstrated in two ways:
Runoff capture BMPs intercept runoff and retain it until it can be completely infiltrated. Detention devices that temporarily detain runoff and then release it slowly over time cannot be used to satisfy a runoff capture requirement. BMPs that are installed to satisfy a runoff capture requirement also will be effective in minimizing the water quality impacts associated with the "first flush" runoff from small storms. Control of the runoff capture design storm will isolate a high proportion of the annual pollutant load that accumulates on impervious surfaces and that would otherwise wash off into receiving streams. However, where specific water quality objectives exist, it is advisable to select a larger design storm than the runoff capture design storm for design of water quality BMPs. §5.3.2 Water Quality Design Storm The percentage of total annual precipitation from large rainfall events is small. Because of their infrequency, the contribution of large storms to the cumulative pollutant load in most watersheds can be ignored. Therefore, when managing runoff for water quality impacts, the control of frequent small rainfall events should be emphasized. Water quality design storms should have the property that 75 to 90 percent of the total annual rainfall volume will occur in storms of equal or smaller magnitude. A method for determining the 75-percent and 90-percent water quality storm volume has been provided in Appendix F. For the five rainfall regions in Pennsylvania, the volume of 75-percent storms, ranges between approximately 0.69 and 1.24 inches. For 90-percent storms the range is 1.13 to 2.04 inches. Typically, the design criteria for water quality BMPs will involve both runoff volume and discharge rate. The runoff retention or detention volume requirement for a BMP can be specified as a percentage of the water quality volume. The water quality volume is the total volume of runoff that will flow to the BMP during the water quality storm. Because of the combined effects of initial abstraction and infiltration, the water quality volume generally will be much smaller than the rainfall volume associated with the water quality design storm. The magnitude of the water quality volume, and therefore the size of water quality BMPs, can be reduced through the use of site layouts that will reduce runoff potential. Most water quality BMPs also will be designed to operate as runoff flows through the devices. In the devices, settlement or filtration efficiency is directly related to current velocities. For these BMPs, the peak runoff rates generated during the water quality design storm should be used for evaluating the treatment efficiency. The design of filter strips, which do not incorporate runoff retention, is based entirely on discharge rate. The value of attenuating runoff peak in advance of most water quality BMPs is apparent. §5.3.3 Runoff Peak Attenuation Design Storm Runoff peak attenuation design storms are the 2-, 5-, 10-, and 25-year return-frequency storms. The design criteria is that postdevelopment peaks should not exceed predevelopment peaks in downstream areas. Peaks can be attenuated in several ways:
In general, the size of detention BMPs can be reduced significantly if site designs incorporate other peak attenuation measures, such as permeable pavement, filters trips, and grass swales. The effectiveness of installing specific BMPs for attenuating peak runoff rates can be evaluated using stormwater simulation methods. Dozens of hydrograph routing techniques have been published (see Maidment, 1993). In general, all of the algorithms, if appropriately applied, will give similar results. However, when using any method the assumptions on which the method is based and the limitations that apply to its use must be understood. Typically peak attenuation design criteria are applied at the immediate downstream boundary of a project area. However, in some developing watersheds, applying the criteria at locations further downstream may be advantageous. Shifts in the arrival times of runoff peaks from tributary drainage areas occasionally can cause unexpected impacts at critical points in a watershed. In watersheds where runoff timing is a critical factor influencing flooding or erosion, local peak control may not achieve the expected benefits. In these cases, it may be worthwhile for municipal governments to consider developing alternative peak control criteria. This will require a watershed-wide approach to stormwater management and planning. Watershed studies show that in certain watersheds runoff peak attenuation is not needed if the runoff capture and water quality design storms are controlled and if the municipalities in the watershed have effective flood control ordinances that prohibit construction in the floodplain. If failure of a BMP during an extreme rainfall event is likely to cause significant downstream impacts, a spillway design flood should be specified. The device should be able to pass the peak runoff rate in a controlled manner and without eroding outfalls and downstream drainages. For sediment basins, constructed treatment wetlands, wet ponds, and dry ponds, the assumption should be that the primary outlets are inoperative when routing the spillway design flood. The most commonly used runoff hydrograph for the spillway design flood is the 100-year return-frequency storm. The relative magnitude of appropriately sized design storm events is summarized in Table 5.1. The overlap in range is due, in part, to the variability of rainfall distribution patterns across the Commonwealth. When selecting design storms, as part of the development of local stormwater management ordinances, regional rainfall characteristics should be taken into account. Table 5.2, which summarizes information developed in Appendix F, may be useful for selecting design storms for use in counties throughout Pennsylvania. §5.4 SEDIMENT POLLUTION CONTROL The appropriate deployment of sediment control BMPs requires a knowledge of the expected nature of sediment releases from disturbed upgradient areas. Essential information for designing the BMPs are:
One potentially useful tool for assessing order-of-magnitude sediment erosion rates for BMP design is the revised universal soil-loss equation (RUSLE). RUSLE is an empirical model that predicts the cumulative annual soil loss from sheet and rill erosion caused by overland flow. The model includes factors that reflect site-specific characteristics of:
Alternatives to RUSLE include process-based computational programs which can simulate sediment delivery rates and evaluate the trap efficiencies of BMPs (e.g., SEDCAD). The settling velocity of sediment particles depends on the specific gravity and the diameter of the particles. The velocity, assuming particles are spherical, can be approximated by using Stokes Law. The time (in minutes) required for a particle to settle 1 foot under quiescent conditions at 68°F is: ts= 0.00935 / ((sg - 1) x D2) where: sg = specific gravity of sediment D = diameter of particle (in mm) A fine sand with a diameter of 0.075 mm requires about 1 minute to settle 1 foot under ideal conditions. Fine silt (0.004 mm), on the other hand, requires 6 hours to settle an equal distance. These calculations point out the importance of knowing the characteristics of the sediment that a BMP will be required to treat. Extremely efficient settling devices, with quiescent settling zones and long detention times, are required for removing silt-size and clay-size particles. Achieving these conditions may be impossible in most BMPs, including sediment traps, filter strips, or conventional sediment basins. However, a variety of BMPs, including wet ponds and sediment basins with skimmers, can achieve high sediment trap efficiencies for silt and clay-sized sediment if they are sized properly. For large projects with stringent discharge requirements, numerical models such as SEDCAD can be used for evaluating the efficiencies of alternative BMP configurations. Although these considerations are most critical for temporary sediment-control measures, most BMPs will accumulate sediment at a slow rate. Even permanently stabilized areas may contribute significant quantities of sediment over extended periods of time. Predictions of sediment accumulation can be used for developing routine maintenance programs for BMPs. The infiltration measures, such as permeable pavement and infiltration trenches, should not be sited where a significant potential for sediment wash-off exists. The use of RUSLE, or a similar model, may help identify problem sites. Suspended solids that contain nutrients and other nonpoint-source pollutants from developing areas frequently are too small or too light to settle out efficiently. BMPs that are based on filtration are best suited for water quality BMPs. These include:
No single BMP will satisfy all of the stormwater control objectives. Therefore, consideration should be given to cost-effective combinations of measures that will achieve the overall objectives for a particular site. Most measures will benefit more than one area. Full advantage should be taken of the multipurpose characteristics of BMPs. The most efficient site designs will result when BMPs are selected in the following order:
The recommended approach is to reintroduce excess runoff from impervious surfaces back into the natural environment as close to generating surfaces as possible. A variety of BMPs are discussed, many of which can be readily integrated into the landscape design of most development projects. This approach discourages site design that would create large uninterrupted impervious surfaces that concentrate stormwater. Ideally, impervious surfaces will be hydrologically divided so that runoff is delivered in smaller volumes that can be accommodated by smaller, less-expensive, and less-obtrusive BMPs. The recommended approach is based on using a combination of stormwater-control devices and measures that are selected for:
In general, constructing large "end-of-pipe" facilities will be counterproductive because of their cost, high maintenance requirements, the amount of land they consume, and their disruption of the landscape. Table 5-3 is a guide for selecting BMPs according to the size of the upgradient drainage area. The table illustrates how the options for selecting BMPs decrease as the size of the tributary drainage increases. In most cases, a larger drainage also will mean higher capital costs and higher maintenance requirements. Maintenance requirements are compared in Table 5-4. As suggested by the table, vegetative BMPs generally will be less costly to maintain. BMPs that use large impoundments, such as dry ponds, wet ponds, and constructed wetlands, will require active programs for operation, maintenance, and periodic repair. Many site-specific conditions may impose restrictions on the use of some BMPs. Table 5-5 can be used as a guide to some of the limitations. In some cases, adverse site conditions can be overcome through careful design. Combinations of BMPs can sometime alleviate difficulties. For instance, high sediment loads that might adversely affect bioretention facilities can be overcome by providing filter strips in upgradient areas. However, improperly located BMPs can lead to poor performance or excessive maintenance requirements. Runoff can be made to flow downhill through a connected series of BMPs. The deployment of BMPs should conform to and, as necessary, amplify natural features of the landscape, such as drainage swales, terraces, and depressions. Constructed wetlands and riparian forest buffers can be the last measure in the "runoff treatment train" before storm flows are released to surface water. Although most constructed wetlands have low infiltration capacities, they provide quantifiable water quality improvements through filtering and biological uptake and help stabilize base flow to adjacent streams. Development proposals for sites that contain or front on streams should preserve riparian forests where they exist and a forested buffer strip should be planted next to streams if no riparian exists. Runoff from many impervious surfaces has a high potential for introducing water pollution to surface water. Runoff from impervious surfaces should be isolated from other runoff so the pollutants can be treated before the runoff is mixed with less contaminated runoff. Possible water quality concerns include:
The type of control measure selected will depend on the specific contaminants of concern for a particular development (see Table 5-6).
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
| About PACD | Conservation Districts | News | Events | Products & Services | Resources | Employment |
© 2003 Pennsylvania Association of Conservation Districts, Inc.