Michigan; Rain Garden Design and Construction - Michigan Technological University
Content
Rain Garden Design and Construction Guidelines
Rooftop runoff capture for homeowners in suburban, urban, and rural areas
By Nancy-Jeanne Bachmann M.S. Candidate Department of Civil & Environmental Engineering Michigan Technological University www.cee.mtu.edu/sustainable_engineering Prepared for: CE5993: Field Engineering in the Developing World
Copyright © Nancy-Jeanne Bachmann, 2006
Rain Garden Design and Construction Guidelines
2
TABLE OF CONTENTS 1 Introduction.................................................................................................................. 4 1.2 Defined.................................................................................................................. 4 1.3 Need/Application................................................................................................... 4 2 Tools and Materials List............................................................................................... 6 3 Site Selection ............................................................................................................... 6 3.1 Physical Rationale ................................................................................................. 6 3.2 Design ................................................................................................................... 6 4 Infiltration Media ......................................................................................................... 6 4.1 Physical Rationale ................................................................................................. 6 4.2 Design ................................................................................................................... 7 4.2.1 Volume of Infiltration Media .......................................................................... 8 5 Vegetation.................................................................................................................... 8 5.1 Physical Rationale ................................................................................................. 8 5.2 Design ................................................................................................................... 9 6 Sizing......................................................................................................................... 10 6.1 Physical Rationale ............................................................................................... 10 6.2 Design ................................................................................................................. 11 6.3 Design Assumptions ............................................................................................ 12 7 Construction............................................................................................................... 12 7.1 Excavation........................................................................................................... 12 7.2 Berm ................................................................................................................... 13 7.3 Backfill and Compaction of Bioretention Soil...................................................... 13 7.4 Planting ............................................................................................................... 13 7.5 Mulch.................................................................................................................. 13 7.6 Piping to Rain Garden ......................................................................................... 13 8 Design Variations....................................................................................................... 14 8.1 On-Site Location ................................................................................................. 14 8.2 On-Site Soil Infiltration Capacity......................................................................... 14 8.3 Alternative Design Objectives ............................................................................. 14 9 Maintenance............................................................................................................... 19 9.1 Vegetation ........................................................................................................... 19 2.2 Clogging Prevention............................................................................................ 19 10 Final Remarks .......................................................................................................... 19 11 Resources................................................................................................................. 20
Rain Garden Design and Construction Guidelines
3
LIST OF FIGURES Figure 1 Typical rain garden capturing rooftop runoff from the downspout. This photo taken by Roger Bannerman in Dane County, Wisconsin (WDNR, 2002). ........................ 4 Figure 2 Example rain garden sizes, shapes, and positioning for a homeowner’s rooftop runoff. The three areas darkened in black are the rain gardens; dashed lines below each garden indicate relative depth. (PGCM, 2006)................................................................. 5 Figure 5 Root Systems of Prairie Plants as diagrammed by Heidi Natura demonstrating the proportional length of various native prairie plants and turf grass (far left) (1995). The longest roots shown are 15 feet long......................................................................... 9 Figure 6 Demonstration of a curb cut for runoff capture from street. This picture is from a project in Burnsville, MN, designed by Barr Engineering Company (2004). ............... 14 Figure 7 Infiltration and recharge facility for enhanced infiltration designed by Prince George’s County, MD (2006)........................................................................................ 15 Figure 8 Filtration and partial recharge facility designed by Prince George’s County, MD (2006). .......................................................................................................................... 16 Figure 9 Infiltration, filtration, and recharge facility designed by Prince George’s County, MD (2006). ................................................................................................................... 16 Figure 10 A filtration-only bioretention cell designed by Prince George’s County, MD (2006). .......................................................................................................................... 17 Figure 11 Sizing of a bioretention cell designed by Prince George’s County, MD (2006), shown for comparison to the LID Center design. ........................................................... 18 Figure 12 Sizing and specifications of a bioretention cell designed by the LID Center (USEPA, 2003). ............................................................................................................ 18 LIST OF TABLES Table 1 Plant list for rain gardens with silt and sandy soils in the southern Wisconsin region provided by Applied Ecological Services, Inc., Brodhead, WI, and published in the Wisconsin DNR’s Rain Gardens: A how-to manual for homeowners (2003)............ 10
Rain Garden Design and Construction Guidelines
4
1 Introduction Low Impact Development (LID) is an ecologically sensitive design approach to stormwater management. Prince George’s County Department of Environmental Resources, a pioneer in LID design and implementation, has developed a heavily referenced Low Impact Development Design Manual. The manual identifies the goal of LID design to maximize onsite storage and infiltration at the parcel level (PGCM, 1997). Bioinfiltration cells, rain gardens in the vernacular, are a rising component of LID design in suburban and, in some cases, urban United States. Bioinfiltration of stormwater is source control mitigation of overburdened storm sewers. Additional benefits include maintaining the natural hydrologic regime and restoring elements of lost ecological systems. To date, most designs strive to accept a certain initial depth of any storm, sometimes called the ‘first flush’ (Traver, 2004). 1.2 Defined In the most general sense, bioinfiltration cells are shallow depressions in the soil to which stormwater is directed to maximize infiltration (Figure 1). They are most often mulched and planted with native vegetation that contributes to water capture capacity via evaporation and transpiration.
Figure 1 Typical rain garden capturing rooftop runoff from the downspout. This photo taken by Roger Bannerman in Dane County, Wisconsin (WDNR, 2002).
Several types of bioinfiltration systems exist: bioretention cells, bioinfiltration cells, vegetated biofilters, rain gardens, grass swales, infiltration trenches, buffer strips. Slight variations in design or function arguably justify the need for different nomenclature. Bioretention might include the use of an underdrain, substantial aggregate backfill, or geotextile lining. Vegetated swales are often a linear design providing moderate conveyance, often installed along a roadway. The terms bioinfiltration cell and rain garden will be used interchangeably within this technical brief. 1.3 Need/Application Bioinfiltration has come about due to negative effects of traditional stormwater management systems and resulting state and federal regulations. Stormwater piping networks rapidly convey stormwater away from its source preventing groundwater
Rain Garden Design and Construction Guidelines
5
recharge and potentially causing flooding, erosion, and combined sewer overflows downstream. Traditional detention basins are not designed for volume reduction, and Emerson et al. (2005) found only a 0.3% watershed-wide reduction in peak storm flow according to the modeling of 100 detention basins. Studies like this demonstrate the need to address runoff through source control. The leading regulation driving site-scale, source control stormwater management is the recent addition of Phase II of the National Pollutant Discharge Elimination System (NPDES) to the 1972 Clean Water Act. It requires small municipalities with separate storm sewer systems (MS4s) to address stormwater runoff from their site with the recommended use of “structural BMPs [Best Management Practices] such as grassed swales or porous pavement” (USEPA, 2000). Rooftop runoff, in particular, can be a substantial contributor to municipal storm sewer systems. Pitt et al. (2002) identified one case where directly connected residential roofs constitute an estimated 30-35% of annual runoff volume in the cities of Phoenix, Seattle and Birmingham. Runoff source control by homeowners can be valuable to the overall goal of stormwater runoff and pollution reduction. 1.4 Report Contents This technical brief strives to provide the homeowner with a guide to construction of an effective rain garden for rooftop runoff in suburban, urban, and rural areas (Figure 2). This is a conservative design approach to capture a designated depth of rainfall from each rain event. The design emulates pre-settlement, natural hydrologic conditions. The design is easily adapted to an underdrain system (Section 8.3) for sites with low permeability soils or where water treatment is the main objective.
Figure 2 Example rain garden sizes, shapes, and positioning for a homeowner’s rooftop runoff. The three areas darkened in black are the rain gardens; dashed lines below each garden indicate relative depth. (PGCM, 2006)
This technical brief appropriately references existing, reputable design manuals and strikes a balance between oversimplified designs and technically complex manuals written for governments, municipal planners, and licensed professional engineers. In
Rain Garden Design and Construction Guidelines
6
particular, it strives to incorporate the most recent design recommendations from research published in the scientific literature. 2 Tools and Materials List The following list describes the materials needed for the installation of a rain garden designed using this manual. Characteristics and quantities of certain materials are further specified in related sections. • • • • • • • • Shovel or optional compact excavator (Section 7) ~50/50 sand / soil mixture, soil consisting of a sandy loam or silt (Section 4) Optional soil specific to planted vegetation (Section 4) Mulch (Section 4) Additional fine sand (Section 4) Additional soil (Section 4) A level (Section 7) Native mesic plants (Section 5)
3 Site Selection 3.1 Physical Rationale An ideal site will aid conveyance to the system and contribute to the hydrologic needs (though minimal) of the vegetation during dry periods. 3.2 Design Select a site at least 10 feet from the foundation of the house to prevent water seepage into the foundation (WDNR, 2003). Select a site with a gentle downward slope (but less than 12%) leading away from the house and into the garden, if possible (WDNR, 2003). Determine whether you will pipe the water from the gutter and pipe directly into the garden or whether the water will flow over land before reaching the rain garden. This decision will influence the amount of pipe needed (contributing to the cost), and, therefore, the distance from the house. 4 Infiltration Media 4.1 Physical Rationale Soils in urban settings get compacted during site construction. Soil compaction significantly reduces infiltration rates compared to native soils. For example, Pitt et al. (2002) compared infiltration rates of non-compacted and compacted sandy soils and obtained a reduction from 13 in/hr to 1.4 in/hr due to compaction. For this reason, a noncompacted, moderate to high infiltration media will backfill the bioinfiltration cell. This
Rain Garden Design and Construction Guidelines
7
media also provides water treatment through sedimentation, filtration, sorption1, and precipitation2 (Hseih and Davis, 2005). Settling occurs in the shallow ponding area, filtering occurs through the layers of media, adsorption3 and cation exchange4 occur within the biologically active organic materials in the filter, and phytoremediation5 occurs by the plants (Clar et al., 2004: PGCM, 2006). Total suspended solids (TSS) in the water entering the rain garden are filtered out in this design. TSS are a concern because they can promote clogging of the media, restricting the hydraulic conductivity6 and rate of infiltration (Beach et al., 2005). It is important to obtain and retain a high enough infiltration rate that the rain garden drains before mosquitoes are able to propagate. They require 7 – 12 days to lay and hatch eggs (WDNR, 2003). The USEPA recommends a conservative drainage time of 48 hours or less to prevent propagation of mosquitoes (1999). This constraint is also satisfied in the design. 4.2 Design Hseih and Davis (2005) recommend two possibilities for optimum water treatment. Due to ASCE copyright, they may not be copied here but are accessible at http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JOEEDU0001310 00011001521000001&idtype=cvips&gifs=yes or http://scitation.aip.org/getpdf/servlet/GetPDFServlet?filetype=pdf&id=JOEEDU0001310 00011001521000001&idtype=cvips&prog=normal if access is approved or on page 1530, Figures 4a and 4b of the referenced journal publication. In this technical brief, the designs will be referred to as Hseih and Davis (2005) Figures 4a and 4b. In both designs, the mulch layer should be of high permeability (d107>0.1mm, 0.00039in) and appropriate uniformity (a d60/d10 value less than 4) to filter TSS (Hseih and Davis, 2005). Hseih and Davis Figure 4a demonstrates a single-layer media design for maximum water quality treatment consisting of 20-70% sandy soil (sandy loam texture) mixed with coarse sand (e.g. d10>0.3mm). Plant species requirements will inform the chosen percentage. The suggested depth is 55-75 cm for best water quality treatment. Infiltration rates were
1
Sorption includes several processes through which solutions (e.g. water and chemical constituents) bind to solids including absorption and adsorption. 2 The precipitation of chemical constituents is the separation of the constituent from solution as a solid. 3 Adsorption is the process at the solution-soil interface in which the soil attracts and holds to its surface a solution. 4 Cation exchange is the switching places of cations (positively charged ions) in solution with cations adsorbed to the soil. 5 Phytoremediation is the neutralization or removal of chemical constituents through vegetation. 6 Hydraulic conductivity is a measure of the soil’s ability to transmit water. This varies with respect to soil moisture content and pore water pressure. 7 The designation dx represents the diameter of particles in an aggregate for which at least x percent are that diameter or finer. This number can be provided by the supplier.
Rain Garden Design and Construction Guidelines
8
found to be between 1.2 – 5.4 cm/min at 15 cm water head8. This design requires minimal construction and maintenance as compared to the design in Hseih and Davis Figure 4b. Hseih and Davis (2005) Figure 4b illustrates a multilayer design for maximum water quality treatment that incorporates a vegetation layer for optimum plant survival and a filter layer for treatment. The filter layer serves as a back-up to the initial vegetation layer. The vegetation layer (25 – 30 cm in depth) is specific to the region of installation. Consult a local naturalist or plant nursery for guidance as you obtain from them the plants. The filter layer consists of the same coarse sand (e.g. d10>0.3mm) and sandy loam soil as recommended in the single layer design, but existing in a 50/50 sand/soil ratio in this case. The Hseih and Davis (2005) Figure 4b design requires more construction and maintenance than the single layer design (Hseih and Davis (2005) Figure 4a). In the latter design, total phosphorus, nitrate, and ammonium removal is expected to be greater by about 50%, 3%, and 9%, respectively (Hseih and Davis, 2005). This relative, additional treatment can be weighed against increased design complexity to inform your decision about which design to install. 4.2.1 Volume of Infiltration Media In order to account for minor compaction during construction, obtain 10% more material than that required according to the diagram’s dimensions. 5 Vegetation 5.1 Physical Rationale In general, vegetation helps to maintain the infiltration capacity of the bioretention cell media (Clar et al. 2004). Bioretention plants are also a component of water treatment because they can uptake some nutrients and heavy metals from the media (Hsieh and Davis, 2005). This is referred to as phytoremediation. In general, native plants are preferred over nonnative plants due to their adaptation to the regional climatic trends. The author’s experience comes from the Midwest where mesic9 prairie species are the plants of choice for rain gardens. Prairie species demonstrate the adaptive nature of plants to their native regions. Figure 3 demonstrates the proportional root depth of a variety of native prairie species as compared to turf grass (far left in the diagram). The longest roots shown are 15 feet long. Long roots and large root masses enable the plants to locate water in periods of drought, long after the turf grass has wilted. The plants are adapted to the trends of rain storms
8
Water head refers to the height of water above the point of interest. Due to gravity, water head increases the downward force of water moving into the soil. 9 The term mesic is a designation for an ecosystem’s level of dryness. It falls in the middle of wet and dry (e.g. in decreasing order of average soil moisture content - wet prairie, mesic prairie, and dry prairie).
Rain Garden Design and Construction Guidelines
9
and dry spells characteristic of their native region. The plants’ physiology increases the capacity for evapotranspiration and precludes the need for irrigation. As such, they are ideal for bioinfiltration cells for water quantity control.
Figure 3 Root Systems of Prairie Plants as diagrammed by Heidi Natura demonstrating the proportional length of various native prairie plants and turf grass (far left) (1995). The longest roots shown are 15 feet long.
The effect that native plants have on water quality control is a growing area of study. The diversity that a native matrix of plants can provide is, conceptually, beneficial to water quality. Plant species with cellular level variations are likely to be conducive to uptake of a variety of constituents.
5.2 Design Plant lists for the region of southern Wisconsin were supplied by Applied Ecological Services and published in the Wisconsin Department of Natural Resources’ (WDNR) design manual (Table 1) (2003). This list merely exemplifies the diversity and approximate number of plant species to obtain. For the Midwest region, it is a good starter list that may be presented to a local native mesic prairie plant grower who can augment and adjust the list to fit your ecological niche.
Rain Garden Design and Construction Guidelines
10
Table 1 Plant list for rain gardens with silt and sandy soils in the southern Wisconsin region provided by Applied Ecological Services, Inc., Brodhead, WI, and published in the Wisconsin DNR’s Rain Gardens: A how-to manual for homeowners (2003). SPECIES NAME COMMON NAME Full to Partial Shade Aquilegia canadensis Aster macrophyllus Carex vulpinoidea Eupatorium maculatum Geranium maculatum Phlox divaricata Rudbeckia subtomentosa Schizachyrium scoparium Solidago flexicaulis Tradescantia ohiensis Zizia aurea Full to Partial Sun Aster navae-angliae Carex vulpinoidea Coreopsis lanceolata Eupatorium perfoliatum Euphorbia corollata Liatrix aspera Monarda fistulosa Physostegia virginiana Rudbeckia subtomentosa Schizchyrium scoparium Solidago Riddelli Tradescantia ohiensis Zizia aurea
Wild columbine Big-leaved aster Fox sedge Spotted Joe-Pye weed Wild geranium Woodland phlox Sweet coneflower Little blue stem Zig zag goldenrod Spiderwort Golden Alexander
New England aster Fox sedge Sand coreopsis Boneset Flowering spurge Rough blazing star Wild Bergamot Obedient plant Sweet coneflower Little blue stem Riddell's goldenrod Spiderwort Golden Alexander
For other regions throughout the United States, the Brooklyn Botanic Garden of Brooklyn, New York, provides starter lists of rain garden plants. These lists can be accessed at http://www.bbg.org/gar2/topics/design/2004sp_raingardens.html (BBG, 2006). It is likely that native plant growers in your region are familiar with rain garden installations and will be able to identify the required vegetation to refine these starter lists. (NOTE: It is a misconception that wetland vegetation is used in these installations. Wetland plants would require irrigation every dry season.) 6 Sizing 6.1 Physical Rationale
Rain Garden Design and Construction Guidelines
11
An underdrained swale system with backfill similar to a bioretention cell provides insight into critical characteristics that influence reduction of runoff, storage, and infiltration. Effectiveness depends on the size of the bioretention cell in addition to other contributing factors. The reduction in runoff will depend on the temporal distribution of runoff contributions to the rain garden and the ability of the soil to infiltrate water. The storage and infiltration will also depend largely on the intensity of the storm and the inflow (Barber et al., 2003). The infiltration rates influencing runoff reduction and storage have already been addressed (Sections 4 and 5). This section addresses the sizing. The sizing method used here is a conservative calculation. It is sized to have an aboveground storage capacity equivalent to a determined depth of runoff from the roof area. This design was used by Dietz and Clausen (2006) in a reputable study on rain garden flow and pollutant retention. These authors found it to be compatible with the 1993 version of The Bioretention Manual developed by Prince George’s County, MD (1993). The updated manual continues to be the most referenced manual and a leader internationally in LID design methods (PGCM, 2006). It utilizes the Natural Resource Conservation Service Curve Number to account for sites with variable land-use. The design discussed here assumes one surface type, the rooftop, and allows for simplified calculations suited to the homeowner’s application. 6.2 Design (1) Determine the depth of rainfall desired for capture. There are a few options for determining this value: a. At a minimum, you can use a 0.5-inch depth of rainfall which could be considered capturing the ‘first-flush’ of the rainfall event. Capturing the first-flush is especially important in roadway runoff applications. In rooftop drainage, areas with considerable atmospheric deposition would cause a contaminated first-flush of runoff; b. Access the National Weather Service’s historic precipitation data from the rain gage nearest you. Locate the total depth of each rain storm for the past several years. From this data determine a constant depth of rainfall from each storm that constitutes 80% of the annual runoff. This will be your desired rainfall capture depth; c. Or you may wish to maximize the acreage of your rain garden to capture as much rainfall as is feasible on your lawn. In this case, you would design on this basis. If desired, you can back calculate your theoretical capture volume. (2) Calculate the plan view (bird’s eye view) area of the rooftop which will drain into the bioretention facility. This is easily based on the dimensions of the foundation of the house. (3) Multiply the plan view area of the roof by the desired rainfall capture depth (based on item (1)) to attain the approximate volume of runoff which the bioretention cell will need to accept.
Rain Garden Design and Construction Guidelines
12
(4) Divide by 4 inches the volume of runoff desired for capture. This is the required surface area of your garden. Four inches is determined to be the maximum storage depth10 of your garden. (5) Determine the shape of your rain garden satisfying the calculated surface area requirements. A good shape is approximately twice as long as it is wide for optimum distribution of inflow throughout the rain garden. The longest dimension should be oriented perpendicular to the slope. Using this technique, the intention is that the bioretention cell will infiltrate the desired volume of runoff. Excess runoff will overflow and run down the lawn to the storm sewer system as it would have before bioretention cell installation. 6.3 Design Assumptions This is a conservative design. The design assumes that all of the desired runoff will fill the basin at one time. In reality, infiltration will occur before the desired depth of precipitation has fallen and the rain garden’s capacity will be greater than the volume for which it was designed. The design also makes a reasonable assumption that runoff from the impervious roof is magnitudes larger than the runoff from the short section of pervious lawn between the roof and the bioretention cell. Runoff from the lawn is considered negligible. 7 Construction The following guidelines are for the construction of the two Hseih and Davis (2005) diagrams referenced in Section 4 as Figures 4a and 4b. All specifications are provided below except for those cases where greater detail is provided in previous sections. 7.1 Excavation Excavate to a depth of 125 cm (49 inches, 4.1 feet) measured from the uphill edge of the rain garden. This depth is based on maximum media depths as shown in the Hseih and Davis (2005) diagrams (referenced in Section 4 as Figures 4a and 4b) including 4 inches of above-ground storage. Side walls can be close to vertical to maximize the volume of infiltration media backfill and to simplify the calculations of needed materials. (NOTE: Do NOT dig deeper than 5 feet without competent personnel trained in excavation safety. In 1993 the U.S. Occupational Safety and Health Administration (OSHA) instituted a law requiring trained personnel on site and proper protection of trench walls for trenches greater than 5 feet deep. Protection from collapse includes sheeting/bracing, shoring or sloping. Go to http://www.osha.gov for more information.)
10
The native vegetation referred to in Section 5 will limit the pooling depth to around 4 inches for satisfactory growing conditions (WDNR 2003). Consultation with a local supplier may adjust this depth.
Rain Garden Design and Construction Guidelines
13
If using heavy equipment, keep it outside the perimeter of the bioretention cell if possible. Minimize soil compaction to preserve infiltrative capacity of the soil. 7.2 Berm Create a berm on the outside edge of the downhill portion of the perimeter of the rain garden. The height of the berm should be equal in elevation to the height of the uphill edge and around 10 inches high. Compact the berm firmly; manually is acceptable. This can be executed by jumping on it aggressively. To protect the berm against erosion, it must have very gently sloping sides and some type of cover. Cover the berm with mulch, or plant it with grass or dry-tolerant plant species. If planting, use straw or erosioncontrol mat for stabilization in the interim time before germination (WDNR, 2003). 7.3 Backfill and Compaction of Bioretention Soil Backfill the bioretention cell in successive horizontal layers (lifts) of 12 inches or less with the prescribed soil mixture in Section 4. Compact each lift by supersaturating the entire area of the rain garden (USEPA, 2003). If pooling occurs, wait until water is drained before placing the next lift. Throughout the process, be sure the layers stay level to facilitate even distribution of inflow and maximum infiltration. For best results, place a level on top of a two by four that extends a greater distance across the surface of the garden. 7.4 Planting Plant the plants at a rate of one plant every 1 to 2 feet on center. For more rapid filling in of plant materials, plant them closer to 1 foot apart. Some installations may include shrubs or other plants that require larger spacing. Spacing may be maintained for aesthetic purposes. However, the natural system, including the mesic prairie system, may be characterized by a dense network of vegetation. To maximize natural ecosystem benefits, this density is recommended, though not requisite. 7.5 Mulch Lay the mulch between the plants at the depth detailed in the mulch specifications. This will require careful placement to avoid dismembering or covering the plants. Be sure the final depth between the mulch and the perimeter (unexcavated soil and berm) is 4 inches and the mulch is level throughout. Mulch may be replaced with compost. Some tests show it has a greater capacity to treat pollutants and higher evapotranspiration rates (Barber et al., 2003). Effectively, any combination of mulch and compost may be used. 7.6 Piping to Rain Garden
Rain Garden Design and Construction Guidelines
14
Using flexible, light-weight, corrugated polyethylene pipe, extend the downspout from contributing roof areas to the uphill edge of the rain garden perimeter. The piping may be aligned with room for an extension (a flared end section or an apron) that disperses the water at the perimeter. If necessary (i.e. erosion occurs at the inlet to the garden), small cobble may be used at the inlet to dissipate the force of the water entering the rain garden. 8 Design Variations 8.1 On-Site Location Rain gardens may be located at the end of the yard along the street curb rather than close to the house. This location has two advantages. Assuming a downward sloping yard, it can capture more on-site runoff. In addition, curb cuts could be made to accept, infiltrate, and treat stormwater runoff from the road (Figure 4).
Figure 4 Demonstration of a curb cut for runoff capture from street. This picture is from a project in Burnsville, MN, designed by Barr Engineering Company (2004).
For a successful case study of an entire neighborhood’s project and explanatory pictures, see Barr Engineering Company’s installation in Burnsville, MN, as documented in Land and Water magazine (2004) available at www.landandwater.com. 8.2 On-Site Soil Infiltration Capacity Two simple, low-technology tests may be performed to determine pre-construction infiltration rates (refer to page 9 of the manual mentioned in this section). If on-site soils have infiltration rates greater than 0.5 in/hr you can design and build the rain garden according to the Wisconsin Department of Natural Resources’ manual, Rain Gardens: A How-to Manual for Homeowners (WDNR, 2003). This manual involves minimal excavation, is user friendly, and is commonly referenced among Midwestern, non-profit organizations that encourage these installations. It is available online at http://www.dnr.state.wi.us/org/water/wm/nps/rg/index.htm. 8.3 Alternative Design Objectives
Rain Garden Design and Construction Guidelines
15
If water quality treatment is prioritized over infiltration and groundwater recharge, an underdrain approach is becoming the standard. The drain pipe allows water to be conveyed through the system when infiltration below and surrounding the cell is too slow (Barber et al., 2003). Below is a collection of cross sections and descriptions delineating variations on the design and their respective functions. In all cases (except for the first), the design incorporates a perforated pipe for the underdrain, a cleanout/observation well11, coarse aggregate, and an optional geotextile covering the entire excavated perimeter or some portion of it. Construction may be done independently or with on-site technical aid. Specifications suitable for a contractor or regulatory approval (if, in the rare case, connecting to a storm sewer) may be required. In all cases, the underdrain system is expected to limit ponding to less than half an hour (PGCM, 2006). All pictures and summarized descriptions are drawn from the Bioretention Manual developed by Prince George’s County, MD (2006). The bioretention cell design in Figure 5 facilitates high recharge of groundwater. The manual recommends in-situ soils with infiltration rates of at least 1 in/hr and a depth of at least 2.5 feet for adequate filtration.
Figure 5 Infiltration and recharge facility for enhanced infiltration designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 6 facilitates high filtration and partial recharge of runoff. The underdrain location ensures a desired rate of drainage. Again, the depth is at least 2.5 feet.
11
A cleanout/observation well is a vertical pipe connecting to the underdrain and protruding slightly above the soil surface. It facilitates monitoring how quickly water is transmitted through the system and cleaning out the underdrain from possible build-up of biological materials. (PGCM, 2006)
Rain Garden Design and Construction Guidelines
16
Figure 6 Filtration and partial recharge facility designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 7 is designed for maximum water quality treatment. It is designed to handle higher nutrient loadings by facilitating a fluctuating aerobic/anaerobic12 zone in the layer below the underdrain. The area below the underdrain simultaneously provides a storage area and recharge zone.
Figure 7 Infiltration, filtration, and recharge facility designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 8 is designed for pre-treatment of highly contaminated water before discharge at an outlet pipe. The liner prevents groundwater contamination.
12
Aerobic and anaerobic conditions refer to the presence or absence of oxygen, respectively, within the soil system. The type of bacteria functioning within the soil (contributing to water treatment) is directly related to which condition is present.
Rain Garden Design and Construction Guidelines
17
Figure 8 A filtration-only bioretention cell designed by Prince George’s County, MD (2006).
The Department of Environmental Resources of Prince George’s County, MD, has developed a comprehensive guide (from which the above diagrams and descriptions originated) to understanding and designing a bioretention system that utilizes an underdrain. The design criteria and conditions are specified for implementation in a residential community. The manual provides complete instruction and examples on how to develop the bioretention plans as well as a corresponding grading plan and sediment and erosion control plan. The document identifies issues and responsibilities for the homeowner, developer, designer, and inspector from the concept phase through the maintenance and operation phases of the bioretention cell installation. It also includes guidance for construction, community involvement, maintenance, and additional notes on compaction. It is available on the County’s website at http://www.co.pg.md.us/Government/AgencyIndex/DER/ESD/Bioretention/bioretention. asp.
For purposes of comparison to the design recommended in this technical brief, Figure 9 is a cross section delineating basic dimensions as recommended by Prince George’s County. The two designs are comparable in this regard.
Rain Garden Design and Construction Guidelines
18
Figure 9 Sizing of a bioretention cell designed by Prince George’s County, MD (2006), shown for comparison to the LID Center design.
An additional set of bioretention cell specifications is notable. The non-profit, Low Impact Development Center in Beltsville, MD, has posted on its website a bioretention specification developed through a Cooperative Assistance Agreement under the US EPA Office of Water 104b(3) Program. It is developed for “local governments, planners, and engineers for developing, administering, and incorporating Low Impact Development (LID) into their aquatic resource protection programs” (USEPA, 2003). Figure 10 is the plan designed for the application discussed here. The specifications can be found online at http://www.lowimpactdevelopment.org/epa03/biospec.htm.
Figure 10 Sizing and specifications of a bioretention cell designed by the LID Center (USEPA, 2003).
NOTE: If installing an underdrain according to these design variations, the infiltration media diagrammed in Hseih and Davis (2005) diagrams (referenced in Section 4 as
Rain Garden Design and Construction Guidelines
19
Figures 4a and 4b) can be a more specific designation of the ‘bioretention soil’ as diagrammed in the underdrain cross section (Figure 10). 9 Maintenance 9.1 Vegetation In its first 2-3 years of growth, required maintenance will be intensive with regard to the vegetation. Until a dense network of vegetation is established, weeding non-natives and especially invasive13 plants will be a necessity. Mulch will help prevent weeds but may simultaneously stall expansion of the planted vegetation. Once the matrix is established, it will be an effective defense against the germination of invasives (Vanderpoel, 2003). Smaller rain gardens can be hand-weeded. Spot herbicide application using stateapproved chemicals is an option for treatment. In the case of herbicide application, it should be used sparingly and in dry, calm conditions only. At no point should application occur in standing water. In the case of wind, drift from the spray may stunt the growth of (or kill) downwind plants. 2.2 Clogging Prevention The rain garden should increase in infiltrative capacity during the initial establishment of the root network of the plants due to the resulting organics in the soil and the increased transpiration. However, rain gardens may be susceptible to clogging after a certain period of time, dependent upon the contaminant and solids loading. An option to restore the infiltrative capacity and break up the biomat14 may be to aerate and loosen up the soil annually. This must be done without uprooting or disturbing the plants. In the dormant season, piercing the soil with a pitch fork in the vertical direction (careful not to overturn the soil) may be the best remedy at the lowest risk of damage. Although untested, it may be effective to do this with care throughout the growing season. 10 Final Remarks Adding a rain garden to your property helps to restore the natural hydrology of the area. It reduces contributions to storm sewers, mitigates downstream flooding, and provides an outlet for education and native ecosystem restoration. Rain gardens also contribute aesthetically to your property, raising your property value. Prince George’s County, Maryland, has identified properties with rain gardens as having increased real estate values by up to 20% (PGCM, 2006). In Prairie Crossing subdivision, they roughly estimated that their homes command a 30% premium over other typical nearby communities (Prairie Crossing Information and Sales Center, 2006).
13
Invasive plants are characterized by their aggressive propagation and prevention of native plant establishment. Usually invasives are nonnative and natives are noninvasive. However, this is not always the case. 14 Biomats are formed as a result of biological buildup (or clogging) in a horizontal zone within the soil.
Rain Garden Design and Construction Guidelines
20
To see a community development utilizing comprehensive Low Impact Development techniques such as rain gardens on a large scale, go to www.prairiecrossing.com. 11 Resources Barr Engineering Company. (September/October 2004). “Burnsville Rainwater Gardens.” Land and Water, 48(5), 47. Barber, M. E., King, S. G., Yonge, D. R., Hathorn, W. E. (2003). “Ecology Ditch: A Best Management Practice for Storm Water Runoff Mitigation.” J. Hydrologic Egrg, 8(3), 111. Beach, D.N.H., McCray, J.E., Lowe, K.S., Siegrist, R.L. (2005). “Temporal Changes in Hydraulic Conductivity of Sand Porous Media Biofilters During Wastewater Infiltration Due to Biomat Formation.” J. Hydrology, 311, 230-243. Bouwer, H. (2002). “Artificial Recharge of Groundwater: Hydrogeology and Engineering.” Hydrogeology Journal, 10, 121-142. Brooklyn Botanic Garden (BBG) (2006). “Rain Garden Plants.” http://www.bbg.org/gar2/topics/design/2004sp_raingardens.html Last accessed 12 April 2006. Christianson, R.D., Barfield, B. J., Hayes, J. C., Gasem, K., Brown, G. O. (2004). “Modeling Effectiveness of Bioretention Cells for Control of Stormwater Quantity and Quality.” ASCE Conf. Proc., Critical Transitions in Water and Environmental Resources Management, 37. Clar, M. L., Barfield, B., O’Connor, T. (2004). “BMP Design Guidelines: Vegetative Biofilters.” ASCE Conf. Proc., Critical Transitions in Water and Environmental Resources Management, 66. Dietz, M. E., Clausen, J. C. (2006). “Saturation to Improve Pollutant Retention in a Rain Garden.” Environmental Science and Technology, 40(4), 1335-1340. Emerson, C. H., Welty, C., Traver, R. G. (2005). “Watershed-Scale Evaluation of a System of Storm Water Detention Basins.” J. Hydrologic Engrg., 10(3), 237-242. Hsieh, C-h, Davis, A.P. (2005). “Evaluation and Optimization of Bioretention Media for Treatment of Urban Storm Water Runoff.” J. of Environ. Engrg., 131(11), 1521-1531. Natura, Heidi. 1995. “Root Systems of Prairie Plants.” Available at http://www.livinghabitats.com/ Pitt, R., Chen, S-E, Clark, S. (2002). “Compacted Urban Soils Effects on Infiltration and Bioretention Stormwater Control Designs.” ASCE Conf. Proc., Urban Drainage, 14.
Rain Garden Design and Construction Guidelines
21
Prairie Crossing Information and Sales Center. Email Correspondence. 11 March 2006. Prince George’s County, Maryland (PGCM), Department of Environmental Resources. The Bioretention Manual; Watershed Protection Branch, MD Department of Environmental Protection: Landover, MD, 1993. Prince George’s County, Maryland (PGCM), Department of Environmental Resources. The Bioretention Manual; Watershed Protection Branch, MD Department of Environmental Resources: Largo, MD, 2006. Available at http://www.goprincegeorgescounty.com/government/agencyindex/der/esd/bioretention/bi oretention.asp Prince George’s County, Maryland (PGCM), Department of Environmental Resources. 1997. Low Impact Development Design Manual. Prince George’s County, MD. Traver, R.G. (2004). “Infiltration Strategies for LID.” ASCE Conf. Proc., World Water Congress, Critical Transitions in Water and Environmental Resources Management, 83. United States Environmental Protection Agency (USEPA) (2003). DrainageBioretention Specification. Cooperative Assistance Agreement, Program 104b(3), Office of Water. http://www.lowimpactdevelopment.org/epa03/biospec.htm Last accessed 30 March 2006. United States Environmental Protection Agency (USEPA). (2000, revised 2005). “Stormwater Phase II Final Rule: Small MS4 Stormwater Program Overview.” http://www.epa.gov/npdes/pubs/fact2-0.pdf Last accessed 22 March 2006. United States Environmental Protection Agency (USEPA). (1999). “Storm Water Technology Fact Sheet: Bioretention. EPA 832-F-99-012. Office of Water, Washington, D.C. Vanderpoel, T. (2003). Restoration Co-Chair, Citizens for Conservation, Barrington, IL. Wisconsin Department of Natural Resources and University of Wisconsin-Extension (WDNR) (2002). “Rain Gardens: A Household Way to Provide Water Quality in Your Community.” Board of Regents of the University of Wisconsin System. Wisconsin Department of Natural Resources and University of Wisconsin-Extension (WDNR) (2003). “Rain Gardens: A How-to Manual for Homeowners.” Board of Regents of the University of Wisconsin System.
Rooftop runoff capture for homeowners in suburban, urban, and rural areas
By Nancy-Jeanne Bachmann M.S. Candidate Department of Civil & Environmental Engineering Michigan Technological University www.cee.mtu.edu/sustainable_engineering Prepared for: CE5993: Field Engineering in the Developing World
Copyright © Nancy-Jeanne Bachmann, 2006
Rain Garden Design and Construction Guidelines
2
TABLE OF CONTENTS 1 Introduction.................................................................................................................. 4 1.2 Defined.................................................................................................................. 4 1.3 Need/Application................................................................................................... 4 2 Tools and Materials List............................................................................................... 6 3 Site Selection ............................................................................................................... 6 3.1 Physical Rationale ................................................................................................. 6 3.2 Design ................................................................................................................... 6 4 Infiltration Media ......................................................................................................... 6 4.1 Physical Rationale ................................................................................................. 6 4.2 Design ................................................................................................................... 7 4.2.1 Volume of Infiltration Media .......................................................................... 8 5 Vegetation.................................................................................................................... 8 5.1 Physical Rationale ................................................................................................. 8 5.2 Design ................................................................................................................... 9 6 Sizing......................................................................................................................... 10 6.1 Physical Rationale ............................................................................................... 10 6.2 Design ................................................................................................................. 11 6.3 Design Assumptions ............................................................................................ 12 7 Construction............................................................................................................... 12 7.1 Excavation........................................................................................................... 12 7.2 Berm ................................................................................................................... 13 7.3 Backfill and Compaction of Bioretention Soil...................................................... 13 7.4 Planting ............................................................................................................... 13 7.5 Mulch.................................................................................................................. 13 7.6 Piping to Rain Garden ......................................................................................... 13 8 Design Variations....................................................................................................... 14 8.1 On-Site Location ................................................................................................. 14 8.2 On-Site Soil Infiltration Capacity......................................................................... 14 8.3 Alternative Design Objectives ............................................................................. 14 9 Maintenance............................................................................................................... 19 9.1 Vegetation ........................................................................................................... 19 2.2 Clogging Prevention............................................................................................ 19 10 Final Remarks .......................................................................................................... 19 11 Resources................................................................................................................. 20
Rain Garden Design and Construction Guidelines
3
LIST OF FIGURES Figure 1 Typical rain garden capturing rooftop runoff from the downspout. This photo taken by Roger Bannerman in Dane County, Wisconsin (WDNR, 2002). ........................ 4 Figure 2 Example rain garden sizes, shapes, and positioning for a homeowner’s rooftop runoff. The three areas darkened in black are the rain gardens; dashed lines below each garden indicate relative depth. (PGCM, 2006)................................................................. 5 Figure 5 Root Systems of Prairie Plants as diagrammed by Heidi Natura demonstrating the proportional length of various native prairie plants and turf grass (far left) (1995). The longest roots shown are 15 feet long......................................................................... 9 Figure 6 Demonstration of a curb cut for runoff capture from street. This picture is from a project in Burnsville, MN, designed by Barr Engineering Company (2004). ............... 14 Figure 7 Infiltration and recharge facility for enhanced infiltration designed by Prince George’s County, MD (2006)........................................................................................ 15 Figure 8 Filtration and partial recharge facility designed by Prince George’s County, MD (2006). .......................................................................................................................... 16 Figure 9 Infiltration, filtration, and recharge facility designed by Prince George’s County, MD (2006). ................................................................................................................... 16 Figure 10 A filtration-only bioretention cell designed by Prince George’s County, MD (2006). .......................................................................................................................... 17 Figure 11 Sizing of a bioretention cell designed by Prince George’s County, MD (2006), shown for comparison to the LID Center design. ........................................................... 18 Figure 12 Sizing and specifications of a bioretention cell designed by the LID Center (USEPA, 2003). ............................................................................................................ 18 LIST OF TABLES Table 1 Plant list for rain gardens with silt and sandy soils in the southern Wisconsin region provided by Applied Ecological Services, Inc., Brodhead, WI, and published in the Wisconsin DNR’s Rain Gardens: A how-to manual for homeowners (2003)............ 10
Rain Garden Design and Construction Guidelines
4
1 Introduction Low Impact Development (LID) is an ecologically sensitive design approach to stormwater management. Prince George’s County Department of Environmental Resources, a pioneer in LID design and implementation, has developed a heavily referenced Low Impact Development Design Manual. The manual identifies the goal of LID design to maximize onsite storage and infiltration at the parcel level (PGCM, 1997). Bioinfiltration cells, rain gardens in the vernacular, are a rising component of LID design in suburban and, in some cases, urban United States. Bioinfiltration of stormwater is source control mitigation of overburdened storm sewers. Additional benefits include maintaining the natural hydrologic regime and restoring elements of lost ecological systems. To date, most designs strive to accept a certain initial depth of any storm, sometimes called the ‘first flush’ (Traver, 2004). 1.2 Defined In the most general sense, bioinfiltration cells are shallow depressions in the soil to which stormwater is directed to maximize infiltration (Figure 1). They are most often mulched and planted with native vegetation that contributes to water capture capacity via evaporation and transpiration.
Figure 1 Typical rain garden capturing rooftop runoff from the downspout. This photo taken by Roger Bannerman in Dane County, Wisconsin (WDNR, 2002).
Several types of bioinfiltration systems exist: bioretention cells, bioinfiltration cells, vegetated biofilters, rain gardens, grass swales, infiltration trenches, buffer strips. Slight variations in design or function arguably justify the need for different nomenclature. Bioretention might include the use of an underdrain, substantial aggregate backfill, or geotextile lining. Vegetated swales are often a linear design providing moderate conveyance, often installed along a roadway. The terms bioinfiltration cell and rain garden will be used interchangeably within this technical brief. 1.3 Need/Application Bioinfiltration has come about due to negative effects of traditional stormwater management systems and resulting state and federal regulations. Stormwater piping networks rapidly convey stormwater away from its source preventing groundwater
Rain Garden Design and Construction Guidelines
5
recharge and potentially causing flooding, erosion, and combined sewer overflows downstream. Traditional detention basins are not designed for volume reduction, and Emerson et al. (2005) found only a 0.3% watershed-wide reduction in peak storm flow according to the modeling of 100 detention basins. Studies like this demonstrate the need to address runoff through source control. The leading regulation driving site-scale, source control stormwater management is the recent addition of Phase II of the National Pollutant Discharge Elimination System (NPDES) to the 1972 Clean Water Act. It requires small municipalities with separate storm sewer systems (MS4s) to address stormwater runoff from their site with the recommended use of “structural BMPs [Best Management Practices] such as grassed swales or porous pavement” (USEPA, 2000). Rooftop runoff, in particular, can be a substantial contributor to municipal storm sewer systems. Pitt et al. (2002) identified one case where directly connected residential roofs constitute an estimated 30-35% of annual runoff volume in the cities of Phoenix, Seattle and Birmingham. Runoff source control by homeowners can be valuable to the overall goal of stormwater runoff and pollution reduction. 1.4 Report Contents This technical brief strives to provide the homeowner with a guide to construction of an effective rain garden for rooftop runoff in suburban, urban, and rural areas (Figure 2). This is a conservative design approach to capture a designated depth of rainfall from each rain event. The design emulates pre-settlement, natural hydrologic conditions. The design is easily adapted to an underdrain system (Section 8.3) for sites with low permeability soils or where water treatment is the main objective.
Figure 2 Example rain garden sizes, shapes, and positioning for a homeowner’s rooftop runoff. The three areas darkened in black are the rain gardens; dashed lines below each garden indicate relative depth. (PGCM, 2006)
This technical brief appropriately references existing, reputable design manuals and strikes a balance between oversimplified designs and technically complex manuals written for governments, municipal planners, and licensed professional engineers. In
Rain Garden Design and Construction Guidelines
6
particular, it strives to incorporate the most recent design recommendations from research published in the scientific literature. 2 Tools and Materials List The following list describes the materials needed for the installation of a rain garden designed using this manual. Characteristics and quantities of certain materials are further specified in related sections. • • • • • • • • Shovel or optional compact excavator (Section 7) ~50/50 sand / soil mixture, soil consisting of a sandy loam or silt (Section 4) Optional soil specific to planted vegetation (Section 4) Mulch (Section 4) Additional fine sand (Section 4) Additional soil (Section 4) A level (Section 7) Native mesic plants (Section 5)
3 Site Selection 3.1 Physical Rationale An ideal site will aid conveyance to the system and contribute to the hydrologic needs (though minimal) of the vegetation during dry periods. 3.2 Design Select a site at least 10 feet from the foundation of the house to prevent water seepage into the foundation (WDNR, 2003). Select a site with a gentle downward slope (but less than 12%) leading away from the house and into the garden, if possible (WDNR, 2003). Determine whether you will pipe the water from the gutter and pipe directly into the garden or whether the water will flow over land before reaching the rain garden. This decision will influence the amount of pipe needed (contributing to the cost), and, therefore, the distance from the house. 4 Infiltration Media 4.1 Physical Rationale Soils in urban settings get compacted during site construction. Soil compaction significantly reduces infiltration rates compared to native soils. For example, Pitt et al. (2002) compared infiltration rates of non-compacted and compacted sandy soils and obtained a reduction from 13 in/hr to 1.4 in/hr due to compaction. For this reason, a noncompacted, moderate to high infiltration media will backfill the bioinfiltration cell. This
Rain Garden Design and Construction Guidelines
7
media also provides water treatment through sedimentation, filtration, sorption1, and precipitation2 (Hseih and Davis, 2005). Settling occurs in the shallow ponding area, filtering occurs through the layers of media, adsorption3 and cation exchange4 occur within the biologically active organic materials in the filter, and phytoremediation5 occurs by the plants (Clar et al., 2004: PGCM, 2006). Total suspended solids (TSS) in the water entering the rain garden are filtered out in this design. TSS are a concern because they can promote clogging of the media, restricting the hydraulic conductivity6 and rate of infiltration (Beach et al., 2005). It is important to obtain and retain a high enough infiltration rate that the rain garden drains before mosquitoes are able to propagate. They require 7 – 12 days to lay and hatch eggs (WDNR, 2003). The USEPA recommends a conservative drainage time of 48 hours or less to prevent propagation of mosquitoes (1999). This constraint is also satisfied in the design. 4.2 Design Hseih and Davis (2005) recommend two possibilities for optimum water treatment. Due to ASCE copyright, they may not be copied here but are accessible at http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JOEEDU0001310 00011001521000001&idtype=cvips&gifs=yes or http://scitation.aip.org/getpdf/servlet/GetPDFServlet?filetype=pdf&id=JOEEDU0001310 00011001521000001&idtype=cvips&prog=normal if access is approved or on page 1530, Figures 4a and 4b of the referenced journal publication. In this technical brief, the designs will be referred to as Hseih and Davis (2005) Figures 4a and 4b. In both designs, the mulch layer should be of high permeability (d107>0.1mm, 0.00039in) and appropriate uniformity (a d60/d10 value less than 4) to filter TSS (Hseih and Davis, 2005). Hseih and Davis Figure 4a demonstrates a single-layer media design for maximum water quality treatment consisting of 20-70% sandy soil (sandy loam texture) mixed with coarse sand (e.g. d10>0.3mm). Plant species requirements will inform the chosen percentage. The suggested depth is 55-75 cm for best water quality treatment. Infiltration rates were
1
Sorption includes several processes through which solutions (e.g. water and chemical constituents) bind to solids including absorption and adsorption. 2 The precipitation of chemical constituents is the separation of the constituent from solution as a solid. 3 Adsorption is the process at the solution-soil interface in which the soil attracts and holds to its surface a solution. 4 Cation exchange is the switching places of cations (positively charged ions) in solution with cations adsorbed to the soil. 5 Phytoremediation is the neutralization or removal of chemical constituents through vegetation. 6 Hydraulic conductivity is a measure of the soil’s ability to transmit water. This varies with respect to soil moisture content and pore water pressure. 7 The designation dx represents the diameter of particles in an aggregate for which at least x percent are that diameter or finer. This number can be provided by the supplier.
Rain Garden Design and Construction Guidelines
8
found to be between 1.2 – 5.4 cm/min at 15 cm water head8. This design requires minimal construction and maintenance as compared to the design in Hseih and Davis Figure 4b. Hseih and Davis (2005) Figure 4b illustrates a multilayer design for maximum water quality treatment that incorporates a vegetation layer for optimum plant survival and a filter layer for treatment. The filter layer serves as a back-up to the initial vegetation layer. The vegetation layer (25 – 30 cm in depth) is specific to the region of installation. Consult a local naturalist or plant nursery for guidance as you obtain from them the plants. The filter layer consists of the same coarse sand (e.g. d10>0.3mm) and sandy loam soil as recommended in the single layer design, but existing in a 50/50 sand/soil ratio in this case. The Hseih and Davis (2005) Figure 4b design requires more construction and maintenance than the single layer design (Hseih and Davis (2005) Figure 4a). In the latter design, total phosphorus, nitrate, and ammonium removal is expected to be greater by about 50%, 3%, and 9%, respectively (Hseih and Davis, 2005). This relative, additional treatment can be weighed against increased design complexity to inform your decision about which design to install. 4.2.1 Volume of Infiltration Media In order to account for minor compaction during construction, obtain 10% more material than that required according to the diagram’s dimensions. 5 Vegetation 5.1 Physical Rationale In general, vegetation helps to maintain the infiltration capacity of the bioretention cell media (Clar et al. 2004). Bioretention plants are also a component of water treatment because they can uptake some nutrients and heavy metals from the media (Hsieh and Davis, 2005). This is referred to as phytoremediation. In general, native plants are preferred over nonnative plants due to their adaptation to the regional climatic trends. The author’s experience comes from the Midwest where mesic9 prairie species are the plants of choice for rain gardens. Prairie species demonstrate the adaptive nature of plants to their native regions. Figure 3 demonstrates the proportional root depth of a variety of native prairie species as compared to turf grass (far left in the diagram). The longest roots shown are 15 feet long. Long roots and large root masses enable the plants to locate water in periods of drought, long after the turf grass has wilted. The plants are adapted to the trends of rain storms
8
Water head refers to the height of water above the point of interest. Due to gravity, water head increases the downward force of water moving into the soil. 9 The term mesic is a designation for an ecosystem’s level of dryness. It falls in the middle of wet and dry (e.g. in decreasing order of average soil moisture content - wet prairie, mesic prairie, and dry prairie).
Rain Garden Design and Construction Guidelines
9
and dry spells characteristic of their native region. The plants’ physiology increases the capacity for evapotranspiration and precludes the need for irrigation. As such, they are ideal for bioinfiltration cells for water quantity control.
Figure 3 Root Systems of Prairie Plants as diagrammed by Heidi Natura demonstrating the proportional length of various native prairie plants and turf grass (far left) (1995). The longest roots shown are 15 feet long.
The effect that native plants have on water quality control is a growing area of study. The diversity that a native matrix of plants can provide is, conceptually, beneficial to water quality. Plant species with cellular level variations are likely to be conducive to uptake of a variety of constituents.
5.2 Design Plant lists for the region of southern Wisconsin were supplied by Applied Ecological Services and published in the Wisconsin Department of Natural Resources’ (WDNR) design manual (Table 1) (2003). This list merely exemplifies the diversity and approximate number of plant species to obtain. For the Midwest region, it is a good starter list that may be presented to a local native mesic prairie plant grower who can augment and adjust the list to fit your ecological niche.
Rain Garden Design and Construction Guidelines
10
Table 1 Plant list for rain gardens with silt and sandy soils in the southern Wisconsin region provided by Applied Ecological Services, Inc., Brodhead, WI, and published in the Wisconsin DNR’s Rain Gardens: A how-to manual for homeowners (2003). SPECIES NAME COMMON NAME Full to Partial Shade Aquilegia canadensis Aster macrophyllus Carex vulpinoidea Eupatorium maculatum Geranium maculatum Phlox divaricata Rudbeckia subtomentosa Schizachyrium scoparium Solidago flexicaulis Tradescantia ohiensis Zizia aurea Full to Partial Sun Aster navae-angliae Carex vulpinoidea Coreopsis lanceolata Eupatorium perfoliatum Euphorbia corollata Liatrix aspera Monarda fistulosa Physostegia virginiana Rudbeckia subtomentosa Schizchyrium scoparium Solidago Riddelli Tradescantia ohiensis Zizia aurea
Wild columbine Big-leaved aster Fox sedge Spotted Joe-Pye weed Wild geranium Woodland phlox Sweet coneflower Little blue stem Zig zag goldenrod Spiderwort Golden Alexander
New England aster Fox sedge Sand coreopsis Boneset Flowering spurge Rough blazing star Wild Bergamot Obedient plant Sweet coneflower Little blue stem Riddell's goldenrod Spiderwort Golden Alexander
For other regions throughout the United States, the Brooklyn Botanic Garden of Brooklyn, New York, provides starter lists of rain garden plants. These lists can be accessed at http://www.bbg.org/gar2/topics/design/2004sp_raingardens.html (BBG, 2006). It is likely that native plant growers in your region are familiar with rain garden installations and will be able to identify the required vegetation to refine these starter lists. (NOTE: It is a misconception that wetland vegetation is used in these installations. Wetland plants would require irrigation every dry season.) 6 Sizing 6.1 Physical Rationale
Rain Garden Design and Construction Guidelines
11
An underdrained swale system with backfill similar to a bioretention cell provides insight into critical characteristics that influence reduction of runoff, storage, and infiltration. Effectiveness depends on the size of the bioretention cell in addition to other contributing factors. The reduction in runoff will depend on the temporal distribution of runoff contributions to the rain garden and the ability of the soil to infiltrate water. The storage and infiltration will also depend largely on the intensity of the storm and the inflow (Barber et al., 2003). The infiltration rates influencing runoff reduction and storage have already been addressed (Sections 4 and 5). This section addresses the sizing. The sizing method used here is a conservative calculation. It is sized to have an aboveground storage capacity equivalent to a determined depth of runoff from the roof area. This design was used by Dietz and Clausen (2006) in a reputable study on rain garden flow and pollutant retention. These authors found it to be compatible with the 1993 version of The Bioretention Manual developed by Prince George’s County, MD (1993). The updated manual continues to be the most referenced manual and a leader internationally in LID design methods (PGCM, 2006). It utilizes the Natural Resource Conservation Service Curve Number to account for sites with variable land-use. The design discussed here assumes one surface type, the rooftop, and allows for simplified calculations suited to the homeowner’s application. 6.2 Design (1) Determine the depth of rainfall desired for capture. There are a few options for determining this value: a. At a minimum, you can use a 0.5-inch depth of rainfall which could be considered capturing the ‘first-flush’ of the rainfall event. Capturing the first-flush is especially important in roadway runoff applications. In rooftop drainage, areas with considerable atmospheric deposition would cause a contaminated first-flush of runoff; b. Access the National Weather Service’s historic precipitation data from the rain gage nearest you. Locate the total depth of each rain storm for the past several years. From this data determine a constant depth of rainfall from each storm that constitutes 80% of the annual runoff. This will be your desired rainfall capture depth; c. Or you may wish to maximize the acreage of your rain garden to capture as much rainfall as is feasible on your lawn. In this case, you would design on this basis. If desired, you can back calculate your theoretical capture volume. (2) Calculate the plan view (bird’s eye view) area of the rooftop which will drain into the bioretention facility. This is easily based on the dimensions of the foundation of the house. (3) Multiply the plan view area of the roof by the desired rainfall capture depth (based on item (1)) to attain the approximate volume of runoff which the bioretention cell will need to accept.
Rain Garden Design and Construction Guidelines
12
(4) Divide by 4 inches the volume of runoff desired for capture. This is the required surface area of your garden. Four inches is determined to be the maximum storage depth10 of your garden. (5) Determine the shape of your rain garden satisfying the calculated surface area requirements. A good shape is approximately twice as long as it is wide for optimum distribution of inflow throughout the rain garden. The longest dimension should be oriented perpendicular to the slope. Using this technique, the intention is that the bioretention cell will infiltrate the desired volume of runoff. Excess runoff will overflow and run down the lawn to the storm sewer system as it would have before bioretention cell installation. 6.3 Design Assumptions This is a conservative design. The design assumes that all of the desired runoff will fill the basin at one time. In reality, infiltration will occur before the desired depth of precipitation has fallen and the rain garden’s capacity will be greater than the volume for which it was designed. The design also makes a reasonable assumption that runoff from the impervious roof is magnitudes larger than the runoff from the short section of pervious lawn between the roof and the bioretention cell. Runoff from the lawn is considered negligible. 7 Construction The following guidelines are for the construction of the two Hseih and Davis (2005) diagrams referenced in Section 4 as Figures 4a and 4b. All specifications are provided below except for those cases where greater detail is provided in previous sections. 7.1 Excavation Excavate to a depth of 125 cm (49 inches, 4.1 feet) measured from the uphill edge of the rain garden. This depth is based on maximum media depths as shown in the Hseih and Davis (2005) diagrams (referenced in Section 4 as Figures 4a and 4b) including 4 inches of above-ground storage. Side walls can be close to vertical to maximize the volume of infiltration media backfill and to simplify the calculations of needed materials. (NOTE: Do NOT dig deeper than 5 feet without competent personnel trained in excavation safety. In 1993 the U.S. Occupational Safety and Health Administration (OSHA) instituted a law requiring trained personnel on site and proper protection of trench walls for trenches greater than 5 feet deep. Protection from collapse includes sheeting/bracing, shoring or sloping. Go to http://www.osha.gov for more information.)
10
The native vegetation referred to in Section 5 will limit the pooling depth to around 4 inches for satisfactory growing conditions (WDNR 2003). Consultation with a local supplier may adjust this depth.
Rain Garden Design and Construction Guidelines
13
If using heavy equipment, keep it outside the perimeter of the bioretention cell if possible. Minimize soil compaction to preserve infiltrative capacity of the soil. 7.2 Berm Create a berm on the outside edge of the downhill portion of the perimeter of the rain garden. The height of the berm should be equal in elevation to the height of the uphill edge and around 10 inches high. Compact the berm firmly; manually is acceptable. This can be executed by jumping on it aggressively. To protect the berm against erosion, it must have very gently sloping sides and some type of cover. Cover the berm with mulch, or plant it with grass or dry-tolerant plant species. If planting, use straw or erosioncontrol mat for stabilization in the interim time before germination (WDNR, 2003). 7.3 Backfill and Compaction of Bioretention Soil Backfill the bioretention cell in successive horizontal layers (lifts) of 12 inches or less with the prescribed soil mixture in Section 4. Compact each lift by supersaturating the entire area of the rain garden (USEPA, 2003). If pooling occurs, wait until water is drained before placing the next lift. Throughout the process, be sure the layers stay level to facilitate even distribution of inflow and maximum infiltration. For best results, place a level on top of a two by four that extends a greater distance across the surface of the garden. 7.4 Planting Plant the plants at a rate of one plant every 1 to 2 feet on center. For more rapid filling in of plant materials, plant them closer to 1 foot apart. Some installations may include shrubs or other plants that require larger spacing. Spacing may be maintained for aesthetic purposes. However, the natural system, including the mesic prairie system, may be characterized by a dense network of vegetation. To maximize natural ecosystem benefits, this density is recommended, though not requisite. 7.5 Mulch Lay the mulch between the plants at the depth detailed in the mulch specifications. This will require careful placement to avoid dismembering or covering the plants. Be sure the final depth between the mulch and the perimeter (unexcavated soil and berm) is 4 inches and the mulch is level throughout. Mulch may be replaced with compost. Some tests show it has a greater capacity to treat pollutants and higher evapotranspiration rates (Barber et al., 2003). Effectively, any combination of mulch and compost may be used. 7.6 Piping to Rain Garden
Rain Garden Design and Construction Guidelines
14
Using flexible, light-weight, corrugated polyethylene pipe, extend the downspout from contributing roof areas to the uphill edge of the rain garden perimeter. The piping may be aligned with room for an extension (a flared end section or an apron) that disperses the water at the perimeter. If necessary (i.e. erosion occurs at the inlet to the garden), small cobble may be used at the inlet to dissipate the force of the water entering the rain garden. 8 Design Variations 8.1 On-Site Location Rain gardens may be located at the end of the yard along the street curb rather than close to the house. This location has two advantages. Assuming a downward sloping yard, it can capture more on-site runoff. In addition, curb cuts could be made to accept, infiltrate, and treat stormwater runoff from the road (Figure 4).
Figure 4 Demonstration of a curb cut for runoff capture from street. This picture is from a project in Burnsville, MN, designed by Barr Engineering Company (2004).
For a successful case study of an entire neighborhood’s project and explanatory pictures, see Barr Engineering Company’s installation in Burnsville, MN, as documented in Land and Water magazine (2004) available at www.landandwater.com. 8.2 On-Site Soil Infiltration Capacity Two simple, low-technology tests may be performed to determine pre-construction infiltration rates (refer to page 9 of the manual mentioned in this section). If on-site soils have infiltration rates greater than 0.5 in/hr you can design and build the rain garden according to the Wisconsin Department of Natural Resources’ manual, Rain Gardens: A How-to Manual for Homeowners (WDNR, 2003). This manual involves minimal excavation, is user friendly, and is commonly referenced among Midwestern, non-profit organizations that encourage these installations. It is available online at http://www.dnr.state.wi.us/org/water/wm/nps/rg/index.htm. 8.3 Alternative Design Objectives
Rain Garden Design and Construction Guidelines
15
If water quality treatment is prioritized over infiltration and groundwater recharge, an underdrain approach is becoming the standard. The drain pipe allows water to be conveyed through the system when infiltration below and surrounding the cell is too slow (Barber et al., 2003). Below is a collection of cross sections and descriptions delineating variations on the design and their respective functions. In all cases (except for the first), the design incorporates a perforated pipe for the underdrain, a cleanout/observation well11, coarse aggregate, and an optional geotextile covering the entire excavated perimeter or some portion of it. Construction may be done independently or with on-site technical aid. Specifications suitable for a contractor or regulatory approval (if, in the rare case, connecting to a storm sewer) may be required. In all cases, the underdrain system is expected to limit ponding to less than half an hour (PGCM, 2006). All pictures and summarized descriptions are drawn from the Bioretention Manual developed by Prince George’s County, MD (2006). The bioretention cell design in Figure 5 facilitates high recharge of groundwater. The manual recommends in-situ soils with infiltration rates of at least 1 in/hr and a depth of at least 2.5 feet for adequate filtration.
Figure 5 Infiltration and recharge facility for enhanced infiltration designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 6 facilitates high filtration and partial recharge of runoff. The underdrain location ensures a desired rate of drainage. Again, the depth is at least 2.5 feet.
11
A cleanout/observation well is a vertical pipe connecting to the underdrain and protruding slightly above the soil surface. It facilitates monitoring how quickly water is transmitted through the system and cleaning out the underdrain from possible build-up of biological materials. (PGCM, 2006)
Rain Garden Design and Construction Guidelines
16
Figure 6 Filtration and partial recharge facility designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 7 is designed for maximum water quality treatment. It is designed to handle higher nutrient loadings by facilitating a fluctuating aerobic/anaerobic12 zone in the layer below the underdrain. The area below the underdrain simultaneously provides a storage area and recharge zone.
Figure 7 Infiltration, filtration, and recharge facility designed by Prince George’s County, MD (2006).
The bioretention cell in Figure 8 is designed for pre-treatment of highly contaminated water before discharge at an outlet pipe. The liner prevents groundwater contamination.
12
Aerobic and anaerobic conditions refer to the presence or absence of oxygen, respectively, within the soil system. The type of bacteria functioning within the soil (contributing to water treatment) is directly related to which condition is present.
Rain Garden Design and Construction Guidelines
17
Figure 8 A filtration-only bioretention cell designed by Prince George’s County, MD (2006).
The Department of Environmental Resources of Prince George’s County, MD, has developed a comprehensive guide (from which the above diagrams and descriptions originated) to understanding and designing a bioretention system that utilizes an underdrain. The design criteria and conditions are specified for implementation in a residential community. The manual provides complete instruction and examples on how to develop the bioretention plans as well as a corresponding grading plan and sediment and erosion control plan. The document identifies issues and responsibilities for the homeowner, developer, designer, and inspector from the concept phase through the maintenance and operation phases of the bioretention cell installation. It also includes guidance for construction, community involvement, maintenance, and additional notes on compaction. It is available on the County’s website at http://www.co.pg.md.us/Government/AgencyIndex/DER/ESD/Bioretention/bioretention. asp.
For purposes of comparison to the design recommended in this technical brief, Figure 9 is a cross section delineating basic dimensions as recommended by Prince George’s County. The two designs are comparable in this regard.
Rain Garden Design and Construction Guidelines
18
Figure 9 Sizing of a bioretention cell designed by Prince George’s County, MD (2006), shown for comparison to the LID Center design.
An additional set of bioretention cell specifications is notable. The non-profit, Low Impact Development Center in Beltsville, MD, has posted on its website a bioretention specification developed through a Cooperative Assistance Agreement under the US EPA Office of Water 104b(3) Program. It is developed for “local governments, planners, and engineers for developing, administering, and incorporating Low Impact Development (LID) into their aquatic resource protection programs” (USEPA, 2003). Figure 10 is the plan designed for the application discussed here. The specifications can be found online at http://www.lowimpactdevelopment.org/epa03/biospec.htm.
Figure 10 Sizing and specifications of a bioretention cell designed by the LID Center (USEPA, 2003).
NOTE: If installing an underdrain according to these design variations, the infiltration media diagrammed in Hseih and Davis (2005) diagrams (referenced in Section 4 as
Rain Garden Design and Construction Guidelines
19
Figures 4a and 4b) can be a more specific designation of the ‘bioretention soil’ as diagrammed in the underdrain cross section (Figure 10). 9 Maintenance 9.1 Vegetation In its first 2-3 years of growth, required maintenance will be intensive with regard to the vegetation. Until a dense network of vegetation is established, weeding non-natives and especially invasive13 plants will be a necessity. Mulch will help prevent weeds but may simultaneously stall expansion of the planted vegetation. Once the matrix is established, it will be an effective defense against the germination of invasives (Vanderpoel, 2003). Smaller rain gardens can be hand-weeded. Spot herbicide application using stateapproved chemicals is an option for treatment. In the case of herbicide application, it should be used sparingly and in dry, calm conditions only. At no point should application occur in standing water. In the case of wind, drift from the spray may stunt the growth of (or kill) downwind plants. 2.2 Clogging Prevention The rain garden should increase in infiltrative capacity during the initial establishment of the root network of the plants due to the resulting organics in the soil and the increased transpiration. However, rain gardens may be susceptible to clogging after a certain period of time, dependent upon the contaminant and solids loading. An option to restore the infiltrative capacity and break up the biomat14 may be to aerate and loosen up the soil annually. This must be done without uprooting or disturbing the plants. In the dormant season, piercing the soil with a pitch fork in the vertical direction (careful not to overturn the soil) may be the best remedy at the lowest risk of damage. Although untested, it may be effective to do this with care throughout the growing season. 10 Final Remarks Adding a rain garden to your property helps to restore the natural hydrology of the area. It reduces contributions to storm sewers, mitigates downstream flooding, and provides an outlet for education and native ecosystem restoration. Rain gardens also contribute aesthetically to your property, raising your property value. Prince George’s County, Maryland, has identified properties with rain gardens as having increased real estate values by up to 20% (PGCM, 2006). In Prairie Crossing subdivision, they roughly estimated that their homes command a 30% premium over other typical nearby communities (Prairie Crossing Information and Sales Center, 2006).
13
Invasive plants are characterized by their aggressive propagation and prevention of native plant establishment. Usually invasives are nonnative and natives are noninvasive. However, this is not always the case. 14 Biomats are formed as a result of biological buildup (or clogging) in a horizontal zone within the soil.
Rain Garden Design and Construction Guidelines
20
To see a community development utilizing comprehensive Low Impact Development techniques such as rain gardens on a large scale, go to www.prairiecrossing.com. 11 Resources Barr Engineering Company. (September/October 2004). “Burnsville Rainwater Gardens.” Land and Water, 48(5), 47. Barber, M. E., King, S. G., Yonge, D. R., Hathorn, W. E. (2003). “Ecology Ditch: A Best Management Practice for Storm Water Runoff Mitigation.” J. Hydrologic Egrg, 8(3), 111. Beach, D.N.H., McCray, J.E., Lowe, K.S., Siegrist, R.L. (2005). “Temporal Changes in Hydraulic Conductivity of Sand Porous Media Biofilters During Wastewater Infiltration Due to Biomat Formation.” J. Hydrology, 311, 230-243. Bouwer, H. (2002). “Artificial Recharge of Groundwater: Hydrogeology and Engineering.” Hydrogeology Journal, 10, 121-142. Brooklyn Botanic Garden (BBG) (2006). “Rain Garden Plants.” http://www.bbg.org/gar2/topics/design/2004sp_raingardens.html Last accessed 12 April 2006. Christianson, R.D., Barfield, B. J., Hayes, J. C., Gasem, K., Brown, G. O. (2004). “Modeling Effectiveness of Bioretention Cells for Control of Stormwater Quantity and Quality.” ASCE Conf. Proc., Critical Transitions in Water and Environmental Resources Management, 37. Clar, M. L., Barfield, B., O’Connor, T. (2004). “BMP Design Guidelines: Vegetative Biofilters.” ASCE Conf. Proc., Critical Transitions in Water and Environmental Resources Management, 66. Dietz, M. E., Clausen, J. C. (2006). “Saturation to Improve Pollutant Retention in a Rain Garden.” Environmental Science and Technology, 40(4), 1335-1340. Emerson, C. H., Welty, C., Traver, R. G. (2005). “Watershed-Scale Evaluation of a System of Storm Water Detention Basins.” J. Hydrologic Engrg., 10(3), 237-242. Hsieh, C-h, Davis, A.P. (2005). “Evaluation and Optimization of Bioretention Media for Treatment of Urban Storm Water Runoff.” J. of Environ. Engrg., 131(11), 1521-1531. Natura, Heidi. 1995. “Root Systems of Prairie Plants.” Available at http://www.livinghabitats.com/ Pitt, R., Chen, S-E, Clark, S. (2002). “Compacted Urban Soils Effects on Infiltration and Bioretention Stormwater Control Designs.” ASCE Conf. Proc., Urban Drainage, 14.
Rain Garden Design and Construction Guidelines
21
Prairie Crossing Information and Sales Center. Email Correspondence. 11 March 2006. Prince George’s County, Maryland (PGCM), Department of Environmental Resources. The Bioretention Manual; Watershed Protection Branch, MD Department of Environmental Protection: Landover, MD, 1993. Prince George’s County, Maryland (PGCM), Department of Environmental Resources. The Bioretention Manual; Watershed Protection Branch, MD Department of Environmental Resources: Largo, MD, 2006. Available at http://www.goprincegeorgescounty.com/government/agencyindex/der/esd/bioretention/bi oretention.asp Prince George’s County, Maryland (PGCM), Department of Environmental Resources. 1997. Low Impact Development Design Manual. Prince George’s County, MD. Traver, R.G. (2004). “Infiltration Strategies for LID.” ASCE Conf. Proc., World Water Congress, Critical Transitions in Water and Environmental Resources Management, 83. United States Environmental Protection Agency (USEPA) (2003). DrainageBioretention Specification. Cooperative Assistance Agreement, Program 104b(3), Office of Water. http://www.lowimpactdevelopment.org/epa03/biospec.htm Last accessed 30 March 2006. United States Environmental Protection Agency (USEPA). (2000, revised 2005). “Stormwater Phase II Final Rule: Small MS4 Stormwater Program Overview.” http://www.epa.gov/npdes/pubs/fact2-0.pdf Last accessed 22 March 2006. United States Environmental Protection Agency (USEPA). (1999). “Storm Water Technology Fact Sheet: Bioretention. EPA 832-F-99-012. Office of Water, Washington, D.C. Vanderpoel, T. (2003). Restoration Co-Chair, Citizens for Conservation, Barrington, IL. Wisconsin Department of Natural Resources and University of Wisconsin-Extension (WDNR) (2002). “Rain Gardens: A Household Way to Provide Water Quality in Your Community.” Board of Regents of the University of Wisconsin System. Wisconsin Department of Natural Resources and University of Wisconsin-Extension (WDNR) (2003). “Rain Gardens: A How-to Manual for Homeowners.” Board of Regents of the University of Wisconsin System.
