Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix C Infiltration Testing and Factors of Safety in Support of the Selection and Design of Volume Reduction Approaches (White Paper #1) 1 Introduction .................................................................................................................................... C-1 2 Role of Infiltration Testing in Different Stages of Project Development .................................. C-2 2.1 Site Assessment / Project Planning Phase ............................................................................... C-2 2.2 BMP Design Phase .................................................................................................................. C-3 3 Review of Infiltration Testing and Estimation Approaches Currently in Use by DOTs and Other Agencies ........................................................................................................................................ C-4 4 Guidance for Selecting Infiltration Testing Methods ............................................................... C-10 4.1 Desktop Approaches and Data Correlation Methods ............................................................ C-11 4.2 Surface and Shallow Excavation Methods ............................................................................ C-12 4.3 Deeper Subsurface Tests ....................................................................................................... C-15 5 Specific Considerations for Infiltration Testing ........................................................................ C-21 5.1 Hydraulic Conductivity versus Infiltration Rate versus Percolation Rate ............................. C-21 5.2 Cut and Fill Conditions.......................................................................................................... C-21 5.3 Effects of Direct and Incidental Compaction ........................................................................ C-22 5.4 Temperature Effects on Infiltration Rate ............................................................................... C-23 5.5 Number of Infiltration Tests Needed ..................................................................................... C-23 6 Selecting a Safety Factor .............................................................................................................. C-24 6.1 Site Suitability Considerations for Selection of an Infiltration Factor of Safety ................... C-25 6.2 Design Related Considerations for Selection of an Infiltration Factor of Safety .................. C-26 6.3 Determining Factor of Safety ................................................................................................ C-28 6.4 Implications of a Factor of Safety in BMP Feasibility and Design ....................................... C-29 7 References ..................................................................................................................................... C-31 C-i
1 Introduction Characterization of potential infiltration rates is a critical step in evaluating the degree to which infiltration can be used to reduce stormwater runoff volume. There are numerous methods for measuring or estimating infiltration rates of soils; however not all methods are equally applicable to stormwater facility siting and design. Likewise, the concept of a âfactor of safetyâ has a range of general meanings in engineering design, and has a more specific meaning in its use for stormwater infiltration design and resulting volume reduction. The purpose of this white paper is to synthesize guidance on infiltration rate characterization that is specific to stormwater volume reduction. This white paper is intended to provide guidance to help answer the following questions: 1. How and where does infiltration testing fit into the project development process? 2. What methods are commonly used to assess and measure infiltrative capacity for stormwater applications? 3. What factors should I consider in selecting the most appropriate testing method for my project? 4. Do I need to apply a factor of safety to infiltration rates? If so, how should I select and apply this factor? This white paper is intended to provide an overview of infiltration testing and how it fits into the development process. This paper is based on a review of stormwater guidance documents, focused literature review of key topics, Geosyntecâs design and construction experience, and professional judgment. The paper is not intended to be an exhaustive reference on infiltration testing. It does not attempt to discuss every method for testing, nor is it intended to provide step-by-step procedures for each method. The user is directed to supplemental resources (referenced in this white paper) or other appropriate references for more specific information. Note, that this white paper does not consider other feasibility criteria that may make infiltration infeasible, such as groundwater contamination and geotechnical considerations (these will be covered under separate white papers). In general, infiltration testing should only be conducted if other feasibility criteria have been evaluated and cleared. C-1
2 Role of Infiltration Testing in Different Stages of Project Development In the process of planning and designing infiltration facilities, there are a number of ways that infiltration testing or estimation factors into project development. At the project planning phases, a designer faces the questions: Where within my project area is infiltration potentially feasible? What volume reduction approaches are potentially suitable for my project? If this initial screening returns positive results and volume reduction approaches that include infiltration are selected, then the designer faces further questions at later phases of design: What infiltration rates should I use to design volume reduction facilities? What factor of safety should I apply? 2.1 Site Assessment / Project Planning Phase Site assessment efforts for infiltration potential should ideally be initiated early in the design process so that volume reduction approaches can be incorporated into the project layout as it is developed. At this phase, it may still be possible to adjust the project layout to preserve areas that have good opportunity for infiltration and configure project grading and drainage so that water can be routed to these areas. There are many factors that influence highway design, such that these projects may have less opportunity for adjustment to layout (i.e., alignment) than other types of projects; however adjustments to grading and drainage routing to improve infiltration opportunities may still be possible if this is initiated early in the design process. At this phase of project development, the project team is faced with two key questions: ⢠Where within my project area is infiltration potentially feasible? ⢠What volume reduction approaches are potentially suitable for my project? The amount of information available to answer these questions at the planning phase of project development may vary. For example, at this phase, project planners may have access to extensive geotechnical investigation reports from previous projects, the results of early investigations for the project of interest, and other information, or may be faced with much more limited data such as county soils maps. A key tradeoff exists between the costs of acquiring additional data and adequacy of existing data for planning-level decisions. If too little information is available, then key opportunities for stormwater infiltration may be missed. For example, in conditions with high variability in soils, testing at 1,000-foot spacing may not identify significant areas of permeable soils between borings that have high opportunity for infiltration. Similarly, if too little information is available, the potential for infiltration may be over- stated. For example, in Seattle, the failure of a right-of-way infiltration project was partially attributed to inadequate spatial resolution of infiltration testing data collected at the project planning phase and associated over-estimation of infiltration capacity (Colwell and Tackett, undated). However, conducting infiltration tests can be costly, and there is a practical limit to how much effort can be allocated to this line item in the project budget. Existing guidance has addressed this tradeoff in in a variety of ways (see further detail in Section 3). Some jurisdictions allow the use of simpler testing methods at the project planning phase that are less precise but also less costly. This can allow the planning-level investigation to cover a relatively broad scope with the intent of identifying areas where more intensive investigation will be focused. Other jurisdictions allow projects to rely only on mapped data, such as Natural Resources Conservation Service (NRCS) county soil surveys for planning-level assessment, but it is suggested that basic field screening of How and where does infiltration testing fit into the project development process? C-2
can be useful for early identification of fatal flaws that may not be caught using desktop methods. At the planning phase, it is generally recommended that screening for other feasibility factors, such as groundwater contamination, depth to groundwater, setbacks from wells and structures, and other criteria be applied first, before identifying potential areas for infiltration testing. This approach can improve efficiency by conducting testing in more focused locations after other feasibility factors have already been assessed and cleared. In the absence of specific local guidance, the decision of whether to collect additional data at the project planning phase, and at what resolution, should be based on project-specific factors, such as: ⢠How variable are the soil conditions at the site? Can I reliably interpolate between a fewer number of tests? ⢠What are the project goals relative to volume reduction? How important is it to provide an exhaustive investigation and quantification of infiltration opportunities? Do applicable regulations require a rigorous demonstration that infiltration is infeasible? ⢠How much would infiltration testing cost as a portion of the project budget? Could other design costs potentially be reduced (i.e., conveyance, flood control) if increased budget were allocated to thoroughly investigating infiltration opportunities? Section 3 provides a summary of the planning-level screening methods currently in use by selected DOTs and other agencies, and Section 4 provides additional information related to selecting and applying these methods. 2.2 BMP Design Phase At the BMP design phase, a more detailed and accurate assessment is needed to quantify infiltration rates for each BMP location. When designing a project to meet specific surface runoff volume reduction goals, the rate at which water percolates into underlying soils is a critical design parameter that affects the time it takes for the BMP to drain as well as the amount of storage capacity available to accept runoff from subsequent storm events. Therefore an accurate estimate of design infiltration rates is clearly a critical need. Overestimating infiltrative capacity can result in failed facilities (i.e., facilities that drain more slowly than intended), with subsequent cost implications due to remediation efforts for these facilities. Underestimating infiltrative capacity can result in over-designing infiltration facilities and associated costs. At this phase of project development, the project team is faced with two key questions: ⢠What infiltration rates should I use to design volume reduction facilities? ⢠What factor of safety should I apply? When developing an infiltration testing strategy to establish design infiltration rates for infiltration facilities, a tradeoff exists between the costs of testing and the quality and resolution of the data available to support design decisions. Similar to planning-level screening, there is always the option to collect additional data; however there is a conceptual âpoint of diminishing returnsâ beyond which the cost of acquiring additional data may not necessarily lead to better decision making. Uncertainty that remains in the measured infiltration rate after field testing can, in part, be mitigated through the use of a higher factor of safety for design, as discussed in greater detail in Section 5. As such, there is a tangible economic tradeoff between the cost of acquiring more data and the additional facility construction cost associated with using a higher factor of safety in design (i.e., larger volumes, larger C-3
footprint). Guidance manuals generally grant discretion to the project engineer and plan review agency to use professional judgment to balance this tradeoff. As a general rule, the value obtained from each design-level infiltration test can be improved by: 1. Using a tiered approach for investigation (i.e., planning-level screening in advance of design- level testing) such that more rigorous design-level tests are conducted only in areas where BMPs are likely to be placed. 2. Using proven testing methods that are acceptable to local jurisdictions and provide reliable estimates of infiltration rate. 3. Selecting methods that are applicable for the project conditions. Section 4 provides a summary of commonly used testing methods for establishing design infiltration rates and provides discussion to assist in selecting testing methods. 3 Review of Infiltration Testing and Estimation Approaches Currently in Use by DOTs and Other Agencies Guidance developed by selected state and local agencies was reviewed to provide a summary of the approaches that are currently in use by DOTs and other agencies for measuring and estimating infiltration capacity. Approaches to infiltration testing and estimation vary as summarized in Table 1. While this summary is not exhaustive, it is intended to provide the user with an introduction to the types of requirements and practices currently in place. A discussion of findings from this review is provided below. Greater detail on several of these infiltration testing methods is provided in Section 4. Discussion of Findings Related to Review of Existing Guidance With respect to the use of different methods for project planning versus design, recommendations vary by jurisdiction. Orange County, CA, Portland, OR, Maryland, and Caltrans include specific approaches for planning-level screening versus design, that include the use of simpler tests for planning-level screening versus more comprehensive approaches for design-level investigation. Other jurisdictions allow the use of desktop resources such as NRCS county soil surveys and/or other maps for initial screening, however these methods should always be supplemented with field observations. Still other jurisdictions do not specifically recognize a two-stage approach for planning-level and design-level exploration. The more specific two-stage approach adopted by Orange County and Portland appears to be a function of the regulatory context under which the guidance was developed. In these areas, the project proponent must first evaluate infiltration, and must demonstrate that infiltration is infeasible before considering other options. For example, in Orange County, the municipal stormwater permit specifically requires each project to conduct a ârigorousâ feasibility assessment for infiltration. As such, the respective guidance manuals are more specific about what must be done at the planning phase to determine if infiltration is feasible or not. The Caltrans Infiltration Basin Siting Study (Caltrans, 2003) was undertaken with the specific intent of identifying infiltration opportunities over a broad geographic area, therefore the study deliberately adopted a phased approach for narrowing down the number of potential sites for testing. In contrast, guidance developed by other jurisdictions was developed under a regulatory context in which infiltration is an option, but is not necessarily required. In this context, more freedom is granted to project What methods are commonly used to assess and measure infiltrative capacity? C-4
proponents regarding site assessment to support BMP selection. For example, a designer may still be interested in determining where infiltration is feasible, but the proponent would not be required to demonstrate that they have adequately assessed infiltration opportunities before moving on to other non- infiltration BMP options. Current trends in stormwater regulations are moving toward requiring evaluation of infiltration feasibility. In general, each jurisdiction accepts a number of different testing methods. More than a dozen infiltration testing methods are specified in the seven guidance manuals reviewed. In addition, some jurisdictions allow other tests to be used at the discretion of a project professional. As might be expected, some of these methods are simply variations on similar tests, although there remain a broad range of distinctly different approaches. Six of the seven references reviewed specify that safety factors must be applied to field test results for use in assessing feasibility and/or for design. Approaches for selecting safety factors vary between jurisdictions. Some require a mandatory use of a specific factor of safety and others allow project proponents to select factors of safety based on a number of design considerations. C-5
Table 1. Selected State and Local Government Approaches to Infiltration Testing Jurisdiction Planning Phase Requirements Design Phase Requirements Safety Factors Orange County, CA Small projects may rely solely on regional NRCS soils maps and data that are already available for the project, such as geotechnical investigations, groundwater maps, etc. If Hydrologic Soil Group (HSG) D soils or other severely limiting feasibility constraints are identified (i.e., very shallow groundwater, mapped contaminant plume), then no further investigation is needed to demonstrate infeasibility of infiltration. Larger projects must conduct infiltration measurements at the planning phase unless other factors make infiltration feasible. A âSimple Open Pit Infiltration Testâ is recommended, however any method approved for design-level testing may also be used. At the planning phase, a licensed geotechnical engineer does not necessarily need to conduct the simple open pit test. If initial feasibility screening finds that infiltration is potentially feasible, then the project must conduct detailed testing. Any of the following tests may be used to establish design infiltration rate, under the supervision of the project geotechnical professional: ⢠Open Pit Falling Head ⢠Well Permeameter Method (USBR 7300-89 test) ⢠Percolation Test procedure from Riverside County (with conversion factor) ⢠Double Ring Infiltrometer ⢠Single Ring Infiltrometer ⢠Other methods as approved by project engineer and reviewing agency For planning-level feasibility screening, the infiltration rate from field measurements must be adjusted by a factor of safety of 2.0 as part of screening whether infiltration is feasible. If the adjusted infiltration rate is less than 0.3 in/hr, then infiltration is considered infeasible. For design purposes, a matrix must be used to select the design factor of safety considering site suitability (methods used, soil texture, site variability, depth to groundwater) and design factors (tributary area, level of pretreatment, redundancy of treatment, and compaction) to compute the design factor of safety. This method yields a total factor of safety from 2 to 9, Discretion is granted to the designer and reviewer. Portland, OR Allows âSimplified Approach Open Pit Infiltration Testâ for initial infiltration rate screening. This test can be conducted by a ânonprofessionalâ. Any of the following tests may be used to establish design infiltration rate: ⢠Open pit falling head ⢠Encased falling head (6â single-ring) ⢠Double ring infiltrometer (ASTM D3385) Minimum safety factors depend on testing methods and the type of project: ⢠Open pit falling head: 2 ⢠Encased falling head test: 2 ⢠Double ring infiltrometer: Public facilities: 1 Private facilities: 2 Higher safety factors may be used at the discretion of the engineer and reviewer. C-6
Jurisdiction Planning Phase Requirements Design Phase Requirements Safety Factors Caltrans1 In this study, screening for potential locations with suitable infiltration rates consisted of the following steps: ⢠Conduct desktop analysis, including NRCS Soil Survey hydrologic soil groups and clay content. ⢠Characterize subsurface lithography and depth to groundwater using boreholes; collect continuous cores. ⢠Measure hydraulic conductivity of core samples obtained from borings (laboratory). ⢠Conduct in-hole tests using USBR 7300-89 well permeameter method. ⢠Evaluate other feasibility criteria such as setbacks from roadway and depth to groundwater. For sites identified as higher potential, the study conducted detailed assessment, including: ⢠Conduct in-hole hydraulic conductivity tests using USBR 7300- 89 well permeameter method at higher spatial resolution. At least 4 tests performed at each site. ⢠For some sites, the study recommended additional tests be conducted at the design level, including extended tests conducted over 48 hours with duplicate runs, to confirm design rates. Applied factor of safety of 2.0 to borehole testing results after converting borehole measurements to estimates of vertical hydraulic conductivity. Maryland (including MDOT) Initial infeasibility screening involves one field test per facility, regardless of type or size, or use of previous testing data, such as the following: ⢠On-site septic percolation testing, which can establish initial rate, water table and/or depth to bedrock, ⢠Geotechnical report on the site prepared by a qualified geotechnical consultant, or ⢠NRCS Soil Mapping showing an unsuitable soil group such as a hydrologic group âDâ soil in a low lying area or a Marlboro Clay (expansive). If initial testing yields the finding that probable infiltration rate is greater than 0.52 in/hr, then the project must conduct both of the following tests to establish design infiltration rate: ⢠Dig test pit to evaluate depth to groundwater, depth to bedrock, soil texture, and other factors. ⢠Conduct encased falling head infiltration test (5 inch diameter). The use of lab testing to establish infiltration rates is explicitly prohibited. Factors of safety are not explicitly considered in the manual. C-7
Jurisdiction Planning Phase Requirements Design Phase Requirements Safety Factors Wisconsin (including WisDOT) Allows use of desktop resources based on soil texture to evaluate infiltration potential. Requires field verification of some characteristics. No specific infiltration test required. If a double ring infiltrometer is used, the test must be done per ASTM D3385. Safety factor based on ratio of permeability of various soil horizons in 5 feet below proposed facility bottom elevation. Analysis of groundwater mounding potential must be conducted. New Jersey (including NJDOT) Does not provide specific guidance for planning-phase testing. The design manual specifies the following testing methods for determining design infiltration rate: ⢠Tube permeameter test (laboratory) ⢠Percolation Test ⢠Pit Bailing Test âProcedure given ⢠Basin flooding test for bedrock ⢠Double ring infiltrometer (ASTM 3385) ⢠USBR 7300-89 (Well permeameter test) ⢠Other recognized constant head permeability test Post-construction testing is required to demonstrate that the facility drains within 72 hours with a safety factor of 2. C-8
Jurisdiction Planning Phase Requirements Design Phase Requirements Safety Factors Western Washington (including WSDOT) Specific criteria for planning phase assessment are not included in the Manual. The manualâs simplified approach for sites less than 1 acre allows the use of any of the following: ⢠Large-scale Pilot Infiltration Test (PIT) (100 sq-ft surface area) ⢠Smaller-scale PIT (20 to 32 sq-ft surface area) ⢠Smaller-scale tests such as double-ring or falling head area allowed with appropriate correction factor ⢠For unconsolidated soils, a grain size analysis method may be used. The manual requires a detailed approach for larger tributary areas, including : ⢠Subsurface explorations (test holes or test pits) to a minimum depth below the system. ⢠Continuous sampling to a minimum depth below the system. ⢠One of the following tests: o Large-scale Pilot Infiltration Test (PIT) preferred o Smaller-scale tests such as double-ring or falling head allowed with appropriate correction factor o Grain size analysis allowed to estimate infiltration rate for unconsolidated soils; must collect one test per stratum encountered in borings. ⢠Assessment of âinfiltration receptorâ to evaluate capacity. Three safety factors are multiplied to yield the total safety factor. These factors account for (1) testing methods, (2) system geometry, and (3) potential for clogging. The combined range of safety factors ranges from approximately 2 to 18. For larger projects, a groundwater mounding model must be run to evaluate potential groundwater mounding issues. 1 â As conducted as part of Infiltration Basin Site Selection Study, Volume I, Report No. CTSW-RT-03-025 (Caltrans, 2003). C-9
4 Guidance for Selecting Infiltration Testing Methods In order to select an infiltration testing method, it is important to understand how each test is applied and what specific physical properties the test is designed to measure. Infiltration testing methods vary considerably in these regards. For example, a borehole percolation test is conducted by drilling a borehole, filling a portion of the hole with water, and monitoring the rate of fall of the water. This test directly measures the 3 dimensional flux of water into the walls and bottom of the borehole. An approximate correction is applied to indirectly estimate the vertical hydraulic conductivity from the results of the borehole test. In contrast, a double-ring infiltrometer test is conducted from the ground surface and is intended to provide a direct estimate of vertical (one-dimensional) infiltration rate at this point. Both of these methods are applicable under different conditions. Tests can be differentiated based on a three key factors: Scale of test: The testing methods described below range from small-scale point measurements to larger scale methods that inundate up to more than 100 square feet. While the cost of testing at larger scales can be prohibitive due to amount of excavation and water needed, the advantage of larger tests is that these tests tend to be more resistant to error introduced by spatial variability in soil properties. Particularly in soils with high variability (i.e., complex layering, non-uniform consistency), the results of small tests may be biased by localized properties that do not necessarily represent the bulk infiltration rate of the larger area of soils that the infiltration system would overlay. For example, a point measurement in an area overlying a small sand lens could significantly over-predict infiltration rates in comparison to what would be expected in full-scale implementation. Larger scale tests also tend to better approximate the âdimensionalityâ of BMPs, as discussed below. Dimensionality of test: While some testing procedures attempt to constrain the direction of infiltration to one dimension (e.g., double-ring infiltrometers), each testing method tends to include some degree of lateral infiltration and vertical infiltration as a function of its dimensionality. This also tends to be true of infiltration BMPs, which infiltrate water into the surrounding soils through both their bottoms and their side walls, in various proportions. Ideally, a test would be used that approximates the dimensionality of the proposed infiltration system to be constructed. However, for purposes of normalizing testing methods, standardized tests are typically used. In selecting a test, and determining what potential for error it may include, the dimensionality of the test in comparison to the dimensionality of the proposed BMP is important to consider. Elevation of infiltrating surface: Testing should be conducted at and below the elevation of the proposed infiltrating surface. In some cases this may be well below the existing ground surface, which may influence the testing that can be conducted. From a practical perspective, it is not possible to conduct a double ring infiltrometer test at an elevation far below the ground surface without extensive excavation. However a well permeameter or borehole method would be well suited in this case. The presence of lower permeable materials at or below the bottom of the BMP, such as fine-grained soil (clays and silts) at depth can significantly reduce the infiltration capacity. Surface testing may miss this. This section provides a summary and comparison of the common testing methods currently in use. Table 2 provides a matrix comparison of these methods. How do I determine which methods are most appropriate for my project? C-10
4.1 Desktop Approaches and Data Correlation Methods This section reviews common methods used to evaluate infiltration characteristics based on desktop- available information available, such as GIS data. This section also introduces methods for estimating infiltration properties via correlations with other measurements. 4.1.1 NRCS Soil Survey Maps and Similar NRCS Soil Surveys are generally available nationwide, and provide a wealth of information about the general geographic distribution of soil units, as well as properties of these soil units near the ground surface (http://websoilsurvey.nrcs.usda.gov/app/HomePage.htm) Soil survey information, specifically characterization of hydrologic soil groups (HSGs), soil texture classes, and presence of hydric soils, can provide useful information for quickly evaluating infiltration potential on a broad geographic context. Several of the jurisdictions reviewed in Section 3 allow the use of soil surveys and/or similar datasets for planning-level screening. Geologic maps may also be available for certain areas at similar spatial resolution. These tend to be more appropriate than the NRCS soils surveys for evaluating the properties of geologic formations below the surficial soil layer, and they can also provide useful information for infiltration feasibility screening. However, guidance manuals and studies generally recommend that these types of datasets should be used with care. The FHWA Urban Drainage Design Manual (Federal Highway Administration, 2001) states that âAlthough infiltration rates are published in many county soils reports, it is advised that good field measurements be made to provide better estimates for these parameters.â Similarly, the Orange County (CA) Technical Guidance Document allows the use of soils maps for feasibility screening as an option for small projects, but only on the condition that the soil type is HSG D and the mapped soil type is confirmed through information available at the specific site. Caltrans (2003) found that use of HSG D classification to exclude study locations resulted in excluding locations that may actually have been found to have good properties for infiltration had site testing been conducted. Generally, confirming mapped conditions with available data from the site (e.g., soil borings, observed soil textures, biological indicators) can provide an inexpensive means of improving the reliability of using regional maps. 4.1.2 Grain Size Analysis Hydraulic conductivity can be estimated indirectly from correlations with soil grain size distributions. While this method is approximate, correlations have been relatively well established for some soil conditions. One of the most commonly used correlations between grain size parameters and hydraulic conductivity is the Hazen (1892, 1911) empirical formula (Philips and Kitch, 2011), but a variety of others have been developed. WADOE and WSDOT accept estimates of infiltration rate developed based on soil grain size distribution. Their method, developed from local experience, uses the ASTM soil size distribution test procedure (ASTM D422), which considers the full range of soil particle sizes, to develop soil size distribution curves. An empirical formula was derived (Massmann 2003, and Massmann et al., 2003) to relate the D10, D60 and D90 to the saturated hydraulic conductivity of an unconsolidated soil sample. The D10, D60, and D90 are the grain sizes for which 10 percent, 60 percent and 90 percent of the sample (by weight) is finer. This analysis must be done for each soil layer encountered below the system to a minimum depth. WADOE and WSDOT accept this method only for soils that have not been consolidated by glacial advance. C-11
Philips and Kitch (2011) found that this method did not consistently align with direct measurements. For their test sites, it was found to result in considerably high estimates compared to direct tests at some sites in which the in-situ material was consolidated to some degree. This is expected, given that compaction of soil has been observed to have significant influence on infiltration rates (Pitt et al. 2008; Cedergren, 1997). Several researchers have also noted high sensitivity of soil infiltration rate to the percent of fines (Cedergren 1997; Hinman 2009, and others) which may not be adequately accounted for using regression methods. For these reasons, grain size methods are considered to have limited reliability for estimating infiltration rates. 4.1.3 Cone Penetrometer Testing Hydraulic conductivity can also be estimated indirectly from cone penetrometer testing (CPT). A cone penetrometer test involves advancing a small probe into the soil and measuring the relative resistance encountered by the probe as it is advanced. The signal returned from this test can be interpreted to yield estimated soil types and the location of key transitions between soil layers. Correlations have also been developed between CPT data and hydraulic conductivity (Lunne et al. 1997). Philips and Kitch (2011) found this method to be highly variable compared to direct measurement. Additional field experience with these methods has not been identified. In general, this method may be useful as an initial planning tool, but does not appear to be reliable for decision making in most cases. 4.2 Surface and Shallow Excavation Methods This section describes tests that are conducted at the ground surface or within shallow excavations close to the ground surface. These tests are generally applicable for cases where the bottom of the infiltration system will be near the existing ground surface. They can also be conducted to confirm the results of borehole methods after excavation/site grading has been completed. 4.2.1 Simple Open Pit Test The Simple Open Pit Test is most appropriate for planning-level screening of infiltration feasibility. Although it is similar to Open Pit Falling Head tests used for establishing a design infiltration rate (see below), the Simple Open Pit Test is less rigorous and is generally conducted to a lower standard of care. Portland (OR) and Orange County (CA) allow this test to be conducted by a nonprofessional as part of planning-level screening phase. The Simple Open Pit Test is a falling head test in which a hole at least 2 feet in diameter is filled to a level of 6â above the bottom. Water level is checked and recorded regularly until either an hour has passed or the entire volume has infiltrated. The test is repeated two more times in succession and the rate at which the water level falls in the third test is used as the infiltration rate. This test has the advantage of being inexpensive to conduct. Yet it is believed to be fairly reliable for screening as the dimensions of the test are similar, proportionally, to the dimensions of a typical BMP. The key limitations of this test are that it measures a relatively small area, does not necessarily result in a precise measurement, and may not be uniformly implemented. Source: City of Portland, 2008. Stormwater Management Manual, Appendix F.2. C-12
4.2.2 Open Pit Falling Head Test This test is similar to the Simple Open Pit Test, but covers a larger footprint, includes more specific instructions, returns more precise measurements, and generally should be overseen by a geotechnical professional. Nonetheless, it remains a relatively simple test. To perform this test, a hole is excavated at least 2 feet wide by 4 feet long (larger is preferred) and to a depth of at least 12 inches. The bottom of the hole should be approximately at the depth of the proposed infiltrating surface of the BMP. The hole is presoaked the by filling it with water at least a foot above the soil to be tested and leaving it at least 4 hours (or overnight if clays are present). After pre-soaking, the hole is refilled to a depth of 12 inches and allow it to drain for one hour (2 hours for slower soils), measuring the rate at which the water level drops. The test is then repeated until successive trials yield a result with less than 10 percent change. In comparison to a double-ring infiltrometer, this test has the advantage of measuring infiltration over a larger area and better resembles the dimensionality of a typical small-scale BMP. Because it includes both vertical and lateral infiltration, it should be adjusted to estimate design rates for larger scale BMPs. Source: County of Orange (2011) 4.2.3 Double Ring Infiltrometer Test (ASTM 3385) The Double Ring Infiltrometer was originally developed to estimate the saturated hydraulic conductivity of low permeability materials, such as clay liners for ponds, but has seen significant use in stormwater applications. The most recent revision of this method from 2009 is known as ASTM 3385-09. The testing apparatus is designed with concentric rings that form an inner ring and an annulus between the inner and outer rings. Infiltration from the annulus between the two rings is intended to saturate the soil outside of the inner ring such that infiltration from the inner ring is restricted primarily to the vertical direction. To conduct this test, both the center ring and annulus between the rings are filled with water. There is no pre-wetting of the soil in this test. However, a constant head of 1 to 6 inches is maintained for 6 hours, or until a constant flow rate is established. Both the inner flow rate and annular flow rate are recorded, but if they are different, the inner flow rate should be used. There are a variety of approaches that are used to maintain a constant head on the system, including use of a Mariotte tube, constant level float valves, or manual observation and filling. This test must be conducted at the elevation of the proposed infiltrating surface, therefore application of this test is limited in cases where the infiltration surface is a significant distance below existing grade at the time of testing. This test is generally considered to provide a direct estimate of vertical infiltration rate for the specific point tested and is highly replicable. However, given the small diameter of the inner ring (standard diameter is 12 inches, but it can be larger), this test only measures infiltration rate in a small area. Additionally, given the small quantity of water used in this test compared to larger scale tests, this test may be biased high in cases where the long term infiltration rate is governed by groundwater mounding and the rate at which mounding dissipates (i.e., the capacity of the infiltration receptor). Finally, the added effort and cost of isolating vertical infiltration rate may not necessarily be warranted considering that BMPs typically have a lateral component of infiltration as well. Therefore, while this method has the advantages of being technical rigorous and well standardized, it should not necessarily be assumed to be the most representative test for estimating full-scale infiltration rates. Source: ASTM International (2009) C-13
4.2.4 Single Ring Infiltrometer Test The single ring infiltrometer test is not a standardized ASTM test, however it is a relatively well- controlled test and shares many similarities with the ASTM standard double ring infiltrometer test (ASTM 3385-09). This test is a constant head test using a large ring (preferably greater than 40 inches in diameter) usually driven 12 inches into the soil. Water is ponded above the surface. The rate of water addition is recorded and infiltration rate is determined after the flow rate has stabilized. Water can be added either manually or automatically. The single ring used in this test tends to be larger than the inner ring used in the double ring test. Driving the ring into the ground limits lateral infiltration; however some lateral infiltration is generally considered to occur. Experience in Riverside County (CA) has shown that this test gives results that are close to full-scale infiltration facilities (Riverside County, 2011). This finding is also supported by King Countyâs Surface Water Design Manual (2009). The primary advantages of this test are that it is relatively simple to conduct and has a larger footprint (compared to the double-ring method) and restricts horizontal infiltration and is more standardized (compared to open pit methods). However, it is still a relatively small-scale test and can only be reasonably conducted near the existing ground surface. 4.2.5 Encased Borehole Tests Encased borehole test methods are similar to single ring infiltrometer tests; however they are typically conducted using a smaller diameter encasement (typically 6 to 12 inches) driven into the native soil that is allowed to drain completely for each test. The encasement ensures that water moves primarily in the vertical direction through the soil plug in the encasement. Generally, these tests measure a smaller surface area than single ring infiltrometers and therefore have greater inherent uncertainty related to spatial heterogeneity. However, they may be less expensive to conduct and practical at greater depths below existing grade. The City of Portland Encased Falling Head Test is an example (City of Portland 2008). Similar methods are used in other jurisdictions. 4.2.6 Large-scale Pilot Infiltration Test (PIT) As its name implies, this test is closer in scale to a full-scale infiltration facility. This test was developed by WADOE specifically for stormwater applications. To perform this test, a test pit is excavated with a horizontal surface area of roughly 100 square feet to a depth that allows 3 to 4 feet of ponding above the expected bottom of the infiltration facility. Water is continually pumped into the system to maintain a constant water level (between 3 and 4 feet about the bottom of the pit, but not more than the estimated water depth in the proposed facility) and the flowrate is recorded. The test is continued until the flow rate stabilizes. Infiltration rate is calculated by dividing the flowrate by the surface area of the pit. Similar to other open pit test, this test is known to result in a slight bias high because infiltration also moves laterally through the walls of the pit during the test. WADOE requires a correction factor of 0.75 (factor of safety of 1.33) be applied to results. This test has the advantage of being more resistant to bias from localized soil variability and being more similar to the dimensionality and scale of full-scale BMPs. It is also more likely to detect long term decline in infiltration rates associated with groundwater mounding. As such, it remains the preferred test for establishing design infiltration rates in Western Washington (WADOE 2012). In a comparative evaluation of test methods, this method was found to provide a more reliable estimate of full-scale C-14
infiltration rate than double ring infiltrometer and borehole percolation tests (Philips and Kitch 2011). King Countyâs Surface Water Design Manual (2009) states that large single ring infiltrometer and PIT tests have proven more effective than smaller test methods at matching as-built performance of infiltration facilities. The difficulty encountered in this method is that it requires a larger area be excavated than the other methods, and this in turn requires larger equipment for excavation and a greater supply of water. However, this method should be strongly considered when less information is known about spatial variability of soils and/or a higher degree of certainty in estimated infiltration rates is desired. WADOE (2012) incentivizes the use of this test by allowing a lower safety factor to be applied to testing results in comparison to the safety factors that must be applied to the results of smaller-scale tests. Source: Washington State Department of Ecology, WADOE (2012) 4.2.7 Smaller-scale Pilot Infiltration Test (PIT) The smaller-scale PIT is conducted similarly to the large-scale PIT but involves a smaller excavation, ranging from 20 to 32 square feet instead of 100 square feet for the large-scale PIT, with similar depths. The primary advantage of this test compared to the full-scale PIT is that it requires less excavation volume and less water. It may be more suitable for small-scale distributed infiltration controls where the need to conduct a greater number of tests outweighs the accuracy that must be obtained in each test, and where groundwater mounding is not as likely to be an issue. WADOE establishes a correction factor of 0.5 (factor of safety of 2.0) for this test in comparison to 0.75 for the large-scale PIT to account for a greater fraction of water infiltrating through the walls of the excavation and lower degree of certainty related to spatial variability of soils. 4.3 Deeper Subsurface Tests 4.3.1 Well Permeameter Method (USBR 7300-89) Well permeameter methods were originally developed for purposes of assessing aquifer permeability and associated yield of drinking water wells. This family of tests is most applicable in situations in which infiltration facilities will be placed substantially below existing grade, which limits the use of surface testing methods. In general, this test involves drilling a 6 inch to 8 inch test well to the depth of interest and maintaining a constant head until a constant flow rate has been achieved. Water level is maintained with down-hole floats. The Porchet method or the nomographs provided in the USBR Drainage Manual (U.S. Department of the Interior, Bureau of Reclamation, 1993) are used to convert the measured rate of percolation to an estimate of vertical hydraulic conductivity. A smaller diameter boring may be adequate, however this then requires a different correction factor to account for the increased variability expected. While these tests have applicability in screening level analysis, considerable uncertainty is introduced in the step of converting direct percolation measurements to estimates of vertical infiltration. Additionally, this testing method is prone to yielding erroneous results cases where the vertical horizon of the test intersects with minor lenses of sandy soils that allow water to dissipate laterally at a much greater rate than would be expected in a full-scale facility. To improve the interpretation of this test method, a continuous soil core can be extracted from the bore hole to determine whether thin lenses of material may C-15
be biasing results at the strata where testing is conducted. This boring should also be extended below the depth of the test, following the completion of the test. Source: (U.S. Department of the Interior, Bureau of Reclamation, 1990, 1993) 4.3.2 Borehole Percolation Tests (Various Methods) Borehole percolation tests were originally developed as empirical tests to estimate the capacity of on- site sewage disposal systems (septic system leach fields), but have more recently been adopted into use for evaluating stormwater infiltration. Similar to the well permeameter method, borehole percolation methods primarily measure lateral infiltration into the walls of the boring and are designed for situations in which infiltration facilities will be placed well below current grade. The percolation rate obtained in this test should be converted to an infiltration rate using a technique such as the Porchet method. This test is generally implemented similarly to the USBR Well Permeameter Method. Per the Riverside County Borehole Percolation method, a hole is bored to a depth at least 5 times the borehole radius. The hole is presoaked for 24 hours (or at least 2 hours if sandy soils with no clay). The hole is filled to approximately the anticipated top of the proposed infiltration basin. Rates of fall are measured for 6 hours, refilling each half hour (or 10 minutes for sand). Tests are generally repeated until consistent results are obtained. The same limitations described for the well permeameter method apply to borehole percolation tests, and their applicability is generally limited to initial screening. To improve the interpretation of this test method, a continuous soil core can be extracted from the hole and below the test depth, following testing, to determine whether thin lenses of material may be biasing results at the strata where testing is conducted. Sources: Riverside County Percolation Test (2011), California Test 750 (1986), San Bernardino County Percolation Test (1992); U.S. EPA Falling Head Test (1980). Porchet Method (aka Inverse Auger Hole Method) for Estimating Saturated Hydraulic Conductivity from Borehole Test Results The Porchet Method (or Inverse Auger Hole Method) is used to estimate one-dimensional saturated hydraulic conductivity of soil based on measurements of the rate of fall of water in a borehole, collected during a borehole percolation test (van Hoorn, 1979, Food and Agriculture Organization of the United Nations, 2007). Data should be recorded after the borehole test has stabilized (i.e., drawdown rates do not vary considerably between sequential trials). When the drop in head is relatively small compared to the total height of water, the following simplified conversion equation can be used (Orange County, 2011): ð¼ð¼ð¡ð¡ = âð»ð»(60ðð)âð¡ð¡(ðð + 2ð»ð»ðððððð) Where: âÎHâ is the change in height over the time interval (Ho â Hf). inches âHoâ is the initial height of water at the selected time interval, measured from the bottom of the hole (inches) âHfâ is the final height of water at the selected time interval, measured from the bottom of the hole (inches) C-16
âHavgâ is the average height of water in the hole over the time interval, measured from the bottom of the hole, inches âÎtâ = test interval, minutes ârâ is the test hole radius, inches âItâ is the resulting infiltration rate, inches per hour. When the ÎH is more than 25 percent of the Havg, more detailed equations may be necessary. See Food and Agriculture Organization of the United Nations (2007) for more detailed methods. 4.3.3 Tube Permeameter Test The Tube Permeameter Test provides an option for estimating infiltration rates in cases where the infiltration surface is well below the ground surface and in-situ borehole testing cannot be conducted. The tube permeameter test is a falling head test in which a core sample, 6 inches long between 1.5 and 3 inches in diameter, is taken from the ground to be tested. The sample is presoaked in de-aired water, and then a column of water is allowed to drain down through the sample. While this test may be reliable for planning-level screening, it has a number of limitations. First, this test is subject to bias introduced by localized variability as a result of the narrow diameter and relatively small dimensions of the core. Second, the removal of the sample can disturb the sample and change its infiltration properties. Finally, this method is not practical for non-cohesive samples due to difficulties in obtaining an intact core sample. For non-cohesive soils, it may be more appropriate to obtain a soil sample and remold it for testing. Furthermore, it is less likely that cohesive soil, which is generally fine grained, will provide suitable infiltration characteristics. Note that some jurisdictions, such as Maryland, explicitly prohibit laboratory methods from being used to establish design infiltration rates. Source: New Jersey Department of Environmental Protection (2009) C-17
Table 2. Summary Matrix of Infiltration Rate Estimation and Testing Methods Test Spatial Scale Dimensionality Vertical Strata Suitable for Planning-Level Screening Suitable for Design Infiltration Rate Estimation Methods NRCS Soil Survey Maps Geographic/ landscape scale (regional maps) NA NRCS soil maps generally provide information for multiple strata; generally within 10 to 20 feet of surface. Geologic maps may be more reliable for deeper subsurface properties. Potentially suitable, but should be interpreted with caution. Should be confirmed with site observations if possible. Not generally suitable for design unless a large factor of safety is applied. Grain Size Analysis Point measurement; however it is relatively simple to obtain a large number of measurements. Correlations provide estimates of vertical Ksat. Samples can be obtained from any strata. Potentially suitable for unconsolidated soils only. Reliability can be improved if correlations are derived and validated based on local soil types. Accepted in WA State for small-scale design in unconsolidated soils, with appropriate factor of safety. Other locations should confirm that correlations are applicable to local soils. Cone Penetrometer Testing Point measurement; however relatively simple to obtain a large number of measurements. Correlations provide estimates of vertical Ksat. Continuous through vertical strata. Potentially suitable; reliability improved if correlations are derived and validated locally. Not generally acceptable for design. C-18
Test Spatial Scale Dimensionality Vertical Strata Suitable for Planning-Level Screening Suitable for Design Infiltration Testing Methods Simple Open Pit Test Point measurement (~4 sq-ft) Has a greater component of horizontal flow than expected in most BMPs. Correction should be applied. Applicable for near- surface only. Generally suitable. Not generally accepted as a design- level test. Can be acceptable if conducted as an Open Pit Falling Head test with professional oversight. Open Pit Falling Head Test Larger point measurement (~8 sq-ft; greater preferred) Greater component of horizontal flow than expected in most BMPs. Correction should be applied. Applicable for near- surface only. Suitable with correction for dimensionality. Suitable with correction for dimensionality. Double Ring Infiltrometer Test (ASTM 3385) Point measurement (~1 sq-ft) Relatively true estimate of vertical rates. Applicable for near- surface only. Suitable, but may be cost prohibitive for preliminary screening of a large area. Generally suitable. Single Ring Infiltrometer Test Larger point measurement (~10 sq-ft) Primarily vertical direction, but with some horizontal direction. Applicable for near- surface only. Suitable, but may be cost prohibitive for preliminary screening of a large area. Suitable; generally a preferred option in jurisdictions using this test. Encased Borehole Tests Smaller point measurement (0.5 to 2.0 ft diameter) Primarily vertical direction, but with some horizontal direction. May be more applicable for deeper tests than similar methods. Suitable, but may be cost prohibitive for preliminary screening of a large area. Generally suitable. Large-scale Pilot Infiltration Test (PIT) Extensive measurement (~100 sq-ft) Dimensionality resembles to lateral proportions expected in typical infiltration BMPs; correction still needed. Applicable for near- surface only. Generally cost prohibitive for preliminary screening of a large area. Suitable; generally a preferred option in jurisdictions using this test. C-19
Test Spatial Scale Dimensionality Vertical Strata Suitable for Planning-Level Screening Suitable for Design Smaller-scale Pilot Infiltration Test (PIT) Smaller-scale extensive measurement (~20 to 32 sq-ft) Dimensionality resembles proportions of vertical to lateral expected in typical small-scale BMPs; requires greater correction than large- scale PIT. Applicable for near- surface only. Suitable, but may be cost prohibitive for preliminary screening of a large area. Suitable; generally preferred in jurisdictions using this test. Well Permeameter Method (USBR 7300-89) Point measurement (3 to 8 inch diameter bore) Primarily lateral infiltration; correction required. Subsurface strata. Generally suitable; reliability of this test can be improved by obtaining a continuous core where tests are conducted. May be appropriate in areas of proposed cut where other tests are not possible; ideally should be confirmed with a more direct measurement following excavation. Borehole Percolation Tests (various methods) Point measurement (6 to 12 inch diameter bore) Primarily lateral infiltration; correction required. Can be conducted from shallow sub- surface to deeper subsurface; most applicable for deeper subsurface. Generally suitable; reliability of this test can be improved by obtaining a continuous core where tests are conducted. May be appropriate in areas of proposed cut where other tests are not possible; ideally should be confirmed with a more direct measurement following excavation. Tube Permeameter Test Point measurement (1.5 to 3 inch diameter bore) Vertical only; conducted within tube in laboratory. Samples can be collected from any strata of a boring; most applicable for deep subsurface investigations where in-situ tests are not possible. Limited reliability, should be used only when other methods are not feasible. Not generally accepted. C-20
5 Specific Considerations for Infiltration Testing The following subsections are intended to address specific topics that commonly arise in characterizing infiltration rates. 5.1 Hydraulic Conductivity versus Infiltration Rate versus Percolation Rate A common misunderstanding is that the âpercolation rateâ obtained from a percolation test is equivalent to the âinfiltration rateâ obtained from tests such as a single or double ring infiltrometer test which is equivalent the âsaturated hydraulic conductivityâ. In fact, these terms have different meanings. Saturated hydraulic conductivity (Ksat) is an intrinsic property of a specific soil sample under a given degree of compaction. It is a coefficient in Darcyâs equation (Darcy 1856) that characterizes the flux of water that will occur under a given gradient. The measurement of Ksat in a laboratory test is typically referred to as âpermeabilityâ, which is a function of the density, structure, stratification, fines, and discontinuities of a given sample under given controlled conditions. In contrast, infiltration rate is an empirical observation of the rate of flux of water into a given soil structure under long term ponding conditions. Similarly to permeability, infiltration rate can be limited by a number of factors including the layering of soil, density, discontinuities, and initial moisture content. These factors control how quickly water can move through a soil. However, infiltration rate can also be influenced by mounding of groundwater, and the rate at which water dissipates horizontally below a BMP â both of which describe the âcapacityâ of the âinfiltration receptorâ to accept this water over an extended period. For this reason, an infiltration test should ideally be conducted for a relatively long duration resembling a series of storm events so that the capacity of the infiltration receptor is evaluated as well as the rate at which water can enter the system. Infiltration rates are generally tested with larger diameter holes, pits, or apparatuses intended to enforce a primarily vertical direction of flux. Permeability can be considered to be synonymous with infiltration rate when the infiltration rate is primarily vertical below the BMP and the capacity of the infiltration receptor is not a limiting factor. In contrast to Ksat, permeability, and infiltration rate, percolation is an empirical observation of the flux of water into a certain soil structure, primarily in the lateral direction. Percolation is tested with small diameter holes, and it mostly a lateral phenomenon. The direct measurement yielded by a percolation test tends to overestimate the infiltration rate, except perhaps in cases in which a BMP has similar dimensionality to the borehole, such as a dry well. Adjustment of percolation rates may be made to an infiltration rate using a technique such as the Porchet Method. 5.2 Cut and Fill Conditions Where the proposed infiltration BMP is to be located in a cut condition, the infiltration surface level at the bottom of the BMP might be far below the existing grade. For example, if the infiltration surface of a proposed BMP is to be located at an elevation that is currently beneath 15 feet of planned cut, how can the proposed infiltration surface be tested to establish a design infiltration rate prior to beginning excavation? The question can be addressed in two ways: First, one of the deeper subsurface tests described above can be used to provide a planning-level screening of potential rates at the elevation of the proposed infiltrating surface. These tests can be conducted at depths exceeding 100 feet, therefore are applicable in most cut conditions. Second, the project can commit to further testing using more reliable methods following bulk excavation to refine or adjust infiltration rates, and/or apply higher factors of C-21
safety to borehole methods to account for the inherent uncertainty in these measurements and conversions. If the bottom of a BMP (infiltration surface) is proposed to be located in a fill location, the infiltration surface may not exist prior to grading. How then can the infiltration rate be determined? For example, if a proposed infiltration BMP is to be located with its bottom elevation in 10 feet of fill, how could one reasonably establish an infiltration rate prior to the fill being placed? Because of uncertainty in material properties as well as concerns regarding geotechnical issues, it is common for guidance manuals to prohibit infiltration into fill. However, if the design process allows for a more detailed understanding of fill properties (potentially through a phased approach, discussed below) and includes consideration of potential geotechnical design impacts, it may be possible to identify locations on a project in which infiltration into fill could be safe and effective. There are two types of fills â those that are engineered or documented, and those that are undocumented. Undocumented fills are fills placed without engineering controls or construction quality assurance and are subject to great uncertainty. On the other hand, engineered fill properties can be very well understood, as they are generally placed using construction quality assurance procedures and may have criteria for grain size and fines content. However, for these types fills, infiltration rates may still be quite uncertain due to layering and heterogeneities introduced as part of construction that cannot be precisely controlled. Where possible, infiltration BMPs should be designed such that their infiltrating surface extends into native soils. Additionally, for shallow fill depths, fill material can be selectively graded (i.e., high permeability granular material placed below proposed BMPs) to provide reliable infiltration properties until the infiltrating water reaches native soils. However, in some cases, due to considerable fill depth, the extension of the BMP down to natural soil and/or selective grading of fill material may prove infeasible. In additional, fill material will result in some compaction of now buried native soils potentially reducing their ability to infiltrate. In these cases, because of the uncertainty of fill parameters as described above as well as potential compaction of the native soils, an infiltration BMP may not be feasible. However, if the source of fill material is defined and this material is known to be of a granular nature and that the native soils below is permeable and will not be highly compacted, infiltration through compacted fill materials may still be feasible. In this case, a project phasing approach could be used including the following general steps, (1) collect samples from areas expected to be used as borrow sites for fill activities, (2) remold samples to approximately the proposed degree of compaction and measure the Ksat of remolded samples using laboratory methods, (3) if infiltration rates appear adequate for infiltration, then apply an appropriate factor of safety and use the initial rates for preliminary design, (4) following placement of fill, conduct in-situ testing to refine design infiltration rates and adjust the design as needed; the infiltration rate of native soil below the fill should also be tested at this time to determine if compaction as a result of fill placement has significantly reduced its infiltration rate. The project geotechnical engineer should be involved in decision making whenever infiltration is proposed in the vicinity of engineered fill structures so that potential impacts of infiltration on the strength and stability of fills and pavement structures can be evaluated. 5.3 Effects of Direct and Incidental Compaction It is widely recognized that compaction of soil has a major influence on infiltration rates (Pitt et al. 2008). However, direct (intentional) compaction is an essential aspect of roadway construction, and indirect C-22
compaction (such as by movement of machinery, placement of fill, stockpiling of materials, and foot traffic) can be difficult to avoid in some parts of the project site. Infiltration testing strategies should attempt to measure soils at a degree of compaction that resembles anticipated post-construction conditions. Ideally, infiltration systems should be located outside of areas where direct compaction will be required and should be staked off to minimize incidental compaction from vehicles and stockpiling. For these conditions, no adjustment of test results is needed. However, in some cases, infiltration BMPs will be constructed in areas to be compacted. For these areas, it may be appropriate to include field compaction tests or prepare laboratory samples and conducting infiltration testing to approximate the degree of compaction that will occur in post-construction conditions. Alternatively, testing could be conducted on undisturbed soil, and an additional factor of safety could be applied to account for anticipated infiltration after compaction. To develop a factor of safety associated with incidental compaction, samples could compacted to various degrees of compaction, their hydraulic conductivity measured, and a âresponse curveâ developed to relate the degree of compaction to the hydraulic conductivity of the material. 5.4 Temperature Effects on Infiltration Rate The rate of infiltration through soil is affected by the viscosity of water, which in turn is affected by the temperature of water. As such, infiltration rate is strongly dependent on the temperature of the infiltrating water (Cedergren, 1997). For example, Emerson (2008) found that wintertime infiltration rates below a BMP in Pennsylvania were approximately half their peak summertime rates. As such, it is important to consider the effects of temperature when planning tests and interpreting results. If possible, testing should be conducted at a temperature that approximates the typical runoff temperatures for the site during the times when rainfall occurs. If this is not possible, then the results of infiltration tests should be adjusted to account for the difference between the temperature at the time of testing and the typical temperature of runoff when rainfall occurs. The measured infiltration can be adjusted by the ratio of the viscosity at the test temperature versus the typical temperature when rainfall occurs (Cedergren, 1997), per the following formula:       ï£ ï£« Ã= Typical Test TestTypical KK µ µ Where: KTypical = the typical infiltration rate expected at typical temperatures when rainfall occurs KTest = the infiltration rate measured or estimated under the conditions of the test µTypical = the viscosity of water at the typical temperature expected when rainfall occurs µTest = the viscosity of water at the temperature at which the test was conducted 5.5 Number of Infiltration Tests Needed The heterogeneity inherent in soils implies that all but the smallest proposed infiltration facilities would benefit from infiltration tests in multiple locations. Indeed, several of the jurisdictions surveyed in this C-23
study provide requirements for the number of infiltration tests they require. The number of infiltration tests specified varies considerably by jurisdiction and is generally a matter that is left to the discretion of the designer and plan reviewer. For example, Orange County (2011) and Portland (2008) have adopted the following requirements for land development: ⢠A total of 2 infiltration tests for every 10,000 square feet of lot area available for new or redevelopment (minimum 2 tests per priority project). An additional test for every 10,000 square feet of lot area available for new or redevelopment. ⢠At least one test for any potential street facility. ⢠One test for every 100 lineal feet of infiltration facility. ⢠In general no more than 5 valid tests are required per development, unless more tests would be valuable or necessary (at the discretion of the qualified professional assessing the site, as well as the reviewing agency). ⢠Infiltration testing should be conducted at each proposed facility. ⢠Testing at multiple strata is recommended. These types of criteria are typical of municipal guidance, particularly where there is an emphasis on rigorously identifying infiltration opportunities as well as establishing design infiltration rates, however specific numbers may vary regarding spacing and frequency of testing. Jurisdictions that do not require a rigorous evaluation of opportunities may only require testing at the locations where BMPs are proposed. Western Washington (2012) requires that tests be conducted at each facility, and at each unique strata of soil, but does not specify a minimum number. However, their guidance allows a lower factor of safety to be used if a greater number of tests are conducted. This incentive is also provided by Orange Countyâs guidance. Caltrans (2003) evaluated selected sites by conducting a minimum of four borehole percolation tests per facility. Their report recommended conducting additional tests at some locations to improve confidence in estimates, specifically where variability in test results was greater. This white paper has not attempted to provide general recommendations regarding the number of tests needed. As a general rule, more tests are needed for sites with higher variability in soil properties and situations in which a higher degree of certainty is desired. The number of samples should be at the discretion of qualified professional assessing the site, as well as the reviewing agency, based on the number of samples needed to characterize the site for its intended use. 6 Selecting a Safety Factor Monitoring of actual facility performance has shown that the full-scale infiltration rate can be much lower than the rate measured by small-scale testing (King County Department of Natural Resources and Parks, 2009). Factors such as soil variability and groundwater mounding may be responsible for much of this difference. Additionally, the infiltration rate of BMPs naturally declines between maintenance cycles as the BMP surface becomes occluded and particulates accumulate in the infiltrative layer. Gulliver et al. (2010) provide the following summary: In the past, infiltration structures have been shown to have a relatively short lifespan. Over 50 percent of infiltration systems either partially or completely failed within the first 5 years of operation Should I use a factor of safety for design infiltration rate? C-24
(U.S. EPA. 1999a). In a Maryland study on infiltration trenches (Lindsey et al. 1991), 53 percent were not operating as designed, 36 percent were clogged, and 22 percent showed reduced filtration. In a study of 12 infiltration basins (Galli 1992), none of which had built-in pretreatment systems, all had failed within the first 2 years of operation. Given the known potential for infiltration BMPs to fail over time, an appropriate factor of safety applied to infiltration testing results is strongly recommended. However, under the evolving regulatory context that increasingly requires infiltration to be used where feasible, the concern has been raised that an âartificiallyâ high factor of safety could be inappropriately used by project proponents to demonstrate that infiltration is infeasible where it actually may be feasible. It is recognized that there are competing objectives in the selection of a factor of safety. There is an initial economic incentive to select a lower factor of safety to yield smaller BMP designs. A low factor of safety also allows a broader range of systems to be considered âfeasibleâ in marginal conditions. However, there are both economic and environmental incentives for the use of an appropriate factor of safety to prevent premature failure and substandard performance. The use of an artificially low factor of safety to demonstrate feasibility in the design process is shortsighted in that it does not consider the long- term feasibility of the system. For these reasons, we recommend that careful consideration be given to the selection of a factor of safety. Local jurisdictions generally take the approach of either prescribing factors of safety that must be used or allowing for the discretion of the engineer and plan reviewer. While this white paper is not intended to supplant local regulations or replace good professional judgment, this section presents a recommended thought process for selecting a safety factor. This method was adapted from technical guidance prepared in Orange County and Ventura County, California. This method considers factor of safety to be a function of: ⢠Site suitability considerations, and ⢠Design-related considerations. These factors and the method for using them to compute a safety factor are discussed below. Importantly, this method encourages rigorous site investigation, good pretreatment, and commitments to routine maintenance to provide technically-sound justification for using a lower factor of safety. When selecting a factor of safety, attention should also be given to factors of safety that may be implicit in other aspects of the design such as the precipitation event used, the runoff coefficient of the tributary area, and other factors, as factors of safety can have compounding effects. Additionally, regenerative processes such as the ability of deeper rooted plants to maintain infiltration pathways should be considered. If other design factors include an implicit or explicit factor of safety and/or if significant regenerative processes are provided in the design, then a lower factor of safety may be warranted for infiltration rate. Using an overly conservative factor of safety may result in over-design and associated excessive costs or a risk in rejecting what would be a suitable condition for infiltration. 6.1 Site Suitability Considerations for Selection of an Infiltration Factor of Safety Considerations related to site suitability include: What factors should be considered in selecting and applying a factor of safety? C-25
⢠Soil assessment methods â the site assessment extent (e.g., number of borings, test pits, etc.) and the measurement method used to estimate the short-term infiltration rate. ⢠Predominant soil texture/percent fines â soil texture and the percent of fines can influence the potential for clogging. Finer grained soils may be more susceptible to clogging. ⢠Site soil variability â site with spatially heterogeneous soils (vertically or horizontally) as determined from site investigations are more difficult to estimate average properties for resulting in a higher level of uncertainty associated with initial estimates. ⢠Depth to seasonal high groundwater/impervious layer â groundwater mounding may become an issue during excessively wet conditions where shallow aquifers or shallow clay lenses are present. Table 3. Suitability Assessment Related Considerations for Infiltration Facility Safety Factors Consideration High Concern â 3 points Medium Concern â 2 points Low Concern â 1 point Assessment methods (see explanation below) Use of soil survey maps or simple texture analysis to estimate short-term infiltration rates Use of well permeameter or borehole methods without accompanying continuous boring log Relatively sparse testing with direct infiltration methods Use of well permeameter or borehole methods with accompanying continuous boring log Direct measurement of infiltration area with localized infiltration measurement methods (e.g., infiltrometer) Moderate spatial resolution Direct measurement with localized (i.e., small- scale) infiltration testing methods at relatively high resolution1 or Use of extensive test pit infiltration measurement methods2 Texture Class Silty and clayey soils with significant fines Loamy soils Granular to slightly loamy soils Site soil variability Highly variable soils indicated from site assessment, or Unknown variability Soil borings/test pits indicate moderately homogeneous soils Soil borings/test pits indicate relatively homogeneous soils Depth to groundwater/ impervious layer <5 ft below facility bottom 5-15 ft below facility bottom >15 below facility bottom 1 - Localized (i.e., small scale) testing refers to methods such as the double-ring infiltrometer and borehole tests) 2 - Extensive infiltration testing refers to methods that include excavating a significant portion of the proposed infiltration area, filling the excavation with water, and monitoring drawdown. The excavation should be to the depth of the proposed infiltration surface and ideally be at least 30 to 100 square feet. 6.2 Design-Related Considerations for Selection of an Infiltration Factor of Safety Design-related considerations include: C-26
⢠Expected influent sediment loads and the level of pretreatment â Sediment loading to the infiltration system is a major factor in the rate at which infiltration rates decline and the potential for failure of the facility increases. For areas with expected sediment in runoff, well designed pretreatment should be included to reduce the probability of clogging from high sediment loading. Infiltration facilities in high sediment loading potential areas should be designed with a higher factor of safety. Infiltration facilities designed to capture runoff from relatively clean surfaces such as rooftops are likely to see low sediment loads and therefore may be designed with lower safety factors. In particular, the amount of landscaped area and its vegetation coverage characteristics tributary to an infiltration facility should be considered. For example in arid areas with more soils exposed, open areas draining to infiltration systems may contribute excessive sediments. Also to be considered is whether sanding is employed for winter traction and the type of truck traffic and materials being transported, both of which can contribute to sediment loading. ⢠Compaction during construction â Proper construction oversight is needed during construction to ensure that the bottoms of infiltration facility are not impacted by significant incidental compaction. Facilities that use proper construction practices and oversight need less restrictive safety factors. ⢠Redundancy/resiliency â Does the design include provisions that would allow the system to continue to operate adequately if conditions are different than design? For example, is there an elevated underdrain system to provide a relief for extended surface ponding should underlying infiltration rates be less than designed for? Can the VRA be designed to allow for maintenance to restore lost infiltration capacity? Are plants with deep/active root systems provided to help maintain infiltration rates and soil health? ⢠Storage depth - The storage depth of the VRA is the total equivalent water depth stored, after accounting for pore spaces. VRAs with deeper storage depths tend to have a higher sediment loading per unit area compared shallower VRAs, which may lead to greater clogging potential. They also may experience more significant issues with extended drain times if clogging does occur. In addition to these factors, the influence of dry weather flows should be considered. For intermittent dry weather flows (say periodic irrigation return flows) where algae or biofilm growth would not be an issue, then infiltration of these flows are typically acceptable, except where they would create a nuisance or potentially impact groundwater quality. For continuous dry weather flow infiltration, one would have to consider that in design and consider potential clogging associated with biofilm. Table 4 describes how to evaluate a given project in these areas. Table 4. Design-Related Considerations for Infiltration Facility Safety Factors Consideration High Concern â 3 points Medium Concern â 2 points Low Concern â one point Level of pretreatment/ expected influent sediment loads Limited pretreatment using gross solids removal devices only, such as hydrodynamic separators, racks and screens AND tributary area includes landscaped areas, steep slopes, high traffic areas, road sanding, or any other areas expected to produce high sediment, trash, or debris loads. Good pretreatment with BMPs that mitigate coarse sediments such as vegetated swales AND influent sediment loads from the tributary area are expected to be moderate (e.g., low traffic, mild slopes, stabilized pervious areas, etc.). Excellent pretreatment with BMPs that mitigate fine sediments such as bioretention or media filtration OR sedimentation or facility only treats runoff from relatively clean surfaces, such as rooftops/non- sanded road surfaces. C-27
Table 4. Design-Related Considerations for Infiltration Facility Safety Factors Consideration High Concern â 3 points Medium Concern â 2 points Low Concern â one point Compaction during construction Construction of facility on a compacted site or increased probability of unintended/ indirect compaction. Medium probability of unintended/ indirect compaction. Equipment traffic is effectively restricted from infiltration areas during construction and there is low probability of unintended/ indirect compaction. Redundancy/ resiliency No âbackupâ system is provided; the system design does not allow infiltration rates to be restored relatively easily with maintenance. The system has a backup pathway for treated water to discharge if clogging occurs or infiltration rates can be restored via maintenance. The system has a backup pathway for treated water to discharge if clogging occurs and infiltration rates can be relatively easily restored via maintenance. Effective Storage Depth of VRA Relatively deep profile (>4 feet) Moderate profile (1 to 4 feet) Shallow profile (< 1 ft) 6.3 Determining Factor of Safety The following procedure can be used to estimate an appropriate factor of safety to be applied to the infiltration testing results. When assigning a factor of safety, care should be taken to understand what other factors of safety are implicit in other aspects of the design to avoid incorporating compounding factors of safety that may result in significant over-design. 1. For each consideration shown above, determine whether the consideration is a high, medium, or low concern. 2. For all high concerns in Table 3, assign a factor value of 3, for medium concerns, assign a factor value of 2, and for low concerns assign a factor value of 1. 3. Multiply each of the factors in Table 3 by 0.25 and then add them together. This should yield a number between 1 and 3. 4. For all high concerns in Table 4, assign a factor value of 3, for medium concerns, assign a factor value of 2, and for low concerns assign a factor value of 1. 5. Multiply each of the factors in Table 4 by 0.5 and then add them together. This should yield a number between 1 and 3. 6. Multiply the two safety factors together to get the final combined safety factor. If the combined safety factor is less than 2, then 2 should be used as the safety factor. 7. Divide the tested infiltration rate by the combined safety factor to obtain the adjusted design infiltration rate for use in sizing the infiltration facility. C-28
Note: The minimum combined adjustment factor should not be less than 2.0 and the maximum combined adjustment factor should not exceed 9.0. Worksheet 1 provides a form for documenting this method. 6.4 Implications of a Factor of Safety in BMP Feasibility and Design The above method will provide safety factors in the range of 2 to 9. From a simplified practical perspective, this means that the size of the facility will need to increase in area from 2 to 9 times relative to that which might be used without a safety factor. Clearly, numbers toward the upper end of this range will make all but the best locations prohibitive in land area and cost. In order to make BMPs more feasible and cost effective, steps should be taken to plan and execute the implementation of infiltration BMPs in a way that will reduce the safety factors needed for those projects. A commitment to thorough site investigation, use of effective pretreatment controls, good construction practices, and restoration of the infiltration rates of soils that are damaged by prior compaction should lower the safety factor that should be applied, to help improve the long term reliability of the system and reduce BMP construction cost. While these practices decrease the recommended safety factor, they do not totally mitigate the need to apply a factor of safety. The minimum recommended safety factor of 2.0 is intended to account for the remaining uncertainty and long-term deterioration that cannot be technically mitigated. For projects being designed under a regulatory mandate to conduct a rigorous infiltration feasibility screening and to select infiltration BMPs where feasible, it may be necessary to put an upper cap on the factor of safety that may be used as part of infeasibility screening. For example, in Orange County (CA), a factor of safety of 2.0 must be used for infiltration feasibility screening such that an artificially high factor of safety cannot be used to inappropriately rule out infiltration. If the site passes the feasibility analysis at a factor of safety of 2.0, then infiltration must investigated, but a higher factor of safety may be selected at the discretion of the design engineer. A similar approach may be useful for DOTs under similar regulatory conditions. C-29
Worksheet 1: Factor of Safety and Design Infiltration Rate Worksheet Factor Category Factor Description Assigned Weight (w) Factor Value (v) Product (p) p = w x v A Suitability Assessment Soil assessment methods 0.25 Predominant soil texture 0.25 Site soil variability 0.25 Depth to groundwater / impervious layer 0.25 Suitability Assessment Safety Factor, SA = Σp B Design Level of pretreatment/ expected sediment loads 0.25 Redundancy/resiliency 0.25 Compaction during construction 0.25 Design infiltration depth 0.25 Design Safety Factor, SB = Σp Combined Safety Factor, Stotal= SA x SB Observed Infiltration Rate, inch/hr, Kobserved (corrected for test-specific bias)1 Design Infiltration Rate, in/hr, Kdesign = Kobserved / Stotal Supporting Data Briefly describe infiltration test and provide reference to test forms: C-30
7 References ASTM International. (2009). ASTM Standard D3385-09. Retrieved from http://www.astm.org/Standards/D3385.htm. Caltrans (1986). Method for Determining the Percolation Rate of Soil Using a 6-inch-diameter Test Hole. California Test 750. http://www.dot.ca.gov/hq/esc/sdsee/wwe/documents/Test_750.pdf. Caltrans (June 2003). Infiltration Basin Site Selection Study, Volume I. Report No. CTSW-RT-03- 025. http://www.dot.ca.gov/hq/env/stormwater/special/newsetup/_pdfs/new_technology/CTSW-RT-03- 025/IFB_Final_Report.pdf. Cedergren, H.R. (1997). Seepage, Drainage, and Flow Nets. John Wiley & Sons, 496 p. Colwell, S. and T. Tackett (undated). Ballard Roadside Rain Gardens, Phase 1 â Lessons Learned. http://water.epa.gov/infrastructure/greeninfrastructure/upload/gi_ballardproject.pdf. Orange County (2011). Technical Guidance Document for the Preparation of Conceptual/Preliminary and/or Project Water Quality Management Plans (WQMPs). http://www.ocwatersheds.com/DocmgmtInternet/Download.aspx?id=638. County of San Bernardino (August 1992). Suitability of Lots and Soils for Use of Leachlines or Seepage Pits, Soil Percolation (PERC) Test Report Standards, On-Site Waste Water Disposal System. Darcy, H. (1856). Les fontaines publiques de la Ville de Dijon (The public fountains of the City of Dijon). Trans. Patricia Bobeck. Dalmont, Paris, France. (Kendall/Hunt, 2004) 506 p. Emerson, C.H. (May 2008). Evaluation of Infiltration Practices as a Means to Control Stormwater Runoff. Doctoral dissertation, Villanova University. Food and Agriculture Organization of the United Nations (2007). Guidelines and computer programs for the planning and design of land drainage systems, (W.H. van der Molen, J. MartÃnez Beltrán, and W.J. Ochs, eds.) Irrigation and Drainage Paper 62, Annex 3 Field methods for measuring hydraulic conductivity. ftp://ftp.fao.org/docrep/fao/010/a0975e/a0975e01.pdf. Federal HighwayAdministration (August 2001). Urban Drainage Design Manual. Hydraulic Engineering Circular No. 22, Second Edition Publication No. FHWA-NHI-01-021. U.S. Department of Transportation. http://isddc.dot.gov/OLPFiles/FHWA/010593.pdf. Galli, J. (1992). Analysis of urban stormwater BMP performance and longevity in Prince Georgeâs County, Maryland. Metropolitan Washington Council of Governments, Washington, D.C. Gulliver, J.S., A.J. Erickson, and P.T. Weiss (eds.). (2010). "Stormwater Treatment: Assessment and Maintenance." University of Minnesota, St. Anthony Falls Laboratory. Minneapolis, MN. http://stormwaterbook.safl.umn.edu/ Hazen, A. (1892). Some Physical Properties of Sands and Gravels, With Special Reference to their Use in Filtration. 24th Annual Rep., Massachusetts State Board of Health, Pub. Doc. No. 34, 539â556. Hazen, A. (1911). Discussion of Dams on Sand Foundationsâ by A.C. Koenig. Trans. Am. Soc. Civ. Eng., 73, 199â203. C-31
Hinman, C. (2009). Bioretention Soil Mix Review and Recommendations for Western Washington. Puget Sound Partnership. http://www.psparchives.com/publications/our_work/stormwater/BSMResultsGuidelines%20Final.pdf. King County Department of Natural Resources and Parks (2009). King County, Washington Surface Water Design Manual. Retrieved from http://your.kingcounty.gov/dnrp/library/water-and- land/stormwater/surface-water-design-manual/SWDM-2009.pdf. Lindsey, G., L. Roberts, and W. Page (1991). Storm Water Management Infiltration. Maryland Department of the Environment, Sediment and Storm Water Administration. Lunne, T., Robertson, P. & Powell, J. (1997). Cone Penetration Testing in Geotechnical Practice, E & FN Spon, London, 312 p. Maryland Department of the Environment (2009). Maryland Stormwater Design Manual, Volumes I & II (Effective October 2000, Revised May 2009) http://www.mde.state.md.us/programs/Water/StormwaterManagementProgram/MarylandStormwaterDesi gnManual/Pages/programs/waterprograms/sedimentandstormwater/stormwater_design/index.aspx. Massmann, J., C. Butchart, and S. Stolar, (2003). Infiltration Characteristics, Performance, and Design of Stormwater Facilities, Final Research Report, University of Washington, WSDOT, and in cooperation with FHWA. http://depts.washington.edu/trac/bulkdisk/pdf/483.1.pdf. Massmann, J.W. and L. Johnson (July-August 2001) A set of exercises illustrating flow in porous media, Ground Water, Volume 34(4), 499â503. New Jersey Department of Environmental Protection, Division of Watershed Management (2009). New Jersey Stormwater Best Management Practices Manual (Appendix E: Soils Testing). Retrieved from http://www.njstormwater.org/bmp_manual2.htm. Philips C. and W. Kitch (2011). A review of methods for characterization of site infiltration with design recommendations. California State Polytechnic University-Pomona. http://www.csupomona.edu/~wakitch/arts/Philips_&_Kitch_2011.pdf. Pitt, R., S. Chen, S. Clark, J. Lantrip, and C. Ong (January 2008). Compactionâs Impacts on Urban Stormwater Infiltration, J. Irr. and Drainage Eng. U.S. Department of the Interior, Bureau of Reclamation (1990). âProcedure for Performing Field Permeability Testing by the Well Permeameter Method (USBR 7300-89)â in Earth Manual, Part 2. Materials Engineering Branch Research and Laboratory Services Division, Denver, Colorado. U.S. Department of the Interior, Bureau of Reclamation. (1993). Drainage Manual: A Water Resources Technical Publication. Retrieved from http://www.usbr.gov/pmts/wquality_land/DrainMan.pdf. U.S. EPA (1980). Onsite Wastewater Treatment and Disposal Systems (EPA No. 625/1-80-012). Retrieved from nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=300043XO.txt U.S. EPA (1999). Preliminary data summary of urban storm water best management practices. EPA- 821-R-99-012, U. S. Environmental Protection Agency, Washington, D.C. Van Hoorn, J.W. (1979). Determining hydraulic conductivity with the inversed auger hole and infiltrometer methods. In J. Wesseling, ed. Proceedings of the International Drainage Workshop, ILRI Publication 25, Wageningen, The Netherlands, 150â154. C-32
Washington State Department of Ecology (2012). Stormwater Management Manual for Western Washington - Volume 3: Hydrologic Analysis and Flow Control BMPs. Retrieved from https://fortress.wa.gov/ecy/publications/summarypages/1210030.html. Wisconsin (2012). Storm Water Post-construction Technical Standards. Dates of individual sections vary. Accessed October 5, 2012. http://dnr.wi.gov/topic/stormwater/stormwater_manual.html. C-33
