Effects of Subsurface Drainage on Pavement Performance: Analysis of the SPS-1 and SPS-2 Field Sections (2007)

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.

33 Field Testing Procedure Pilot testing of the procedure and equipment developed for determining the flow rate of water through the subsur- face drainage systems in the SPS-1 and SPS-2 sites was con- ducted at the SPS-1 and SPS-2 sites in Arkansas and at other, non-LTPP sites in Wisconsin. Based on that testing, some improvements were made to the procedure before flow rate testing was conducted at the remaining SPS-1 and SPS-2 sites. The testing procedure took at most 1 hour per test section, or a maximum of 6 hours at SPS-1 sites and a maximum of 4 hours at SPS-2 sites. To accommodate the testing, the state DOT either set up a moving traffic control operation or closed a full outer lane. The testing procedure is described below. Locating and Clearing the Outlets For each of the drained test sections (the six sections with permeable asphalt-treated base layers and edgedrain/outlet systems at each SPS-1 site, and the four sections with perme- able asphalt-treated base layers and edgedrain/outlet systems at each SPS-2 site), the drainage outlets were located. The results from the video inspections conducted earlier were helpful in locating the outlets. In addition, state DOT staff and regional LTPP center representatives knowledgeable about the construction of the test sections were often on site to assist with locating the outlets. At many of the sites, the outlet headwalls were unmarked and obscured by tall vegetation (see Figures 43 through 46, for example). At some sites, the outlet headwalls were also completely covered by dirt, gravel, and other vegetation that had to be dug out with hand tools (see Figures 47 through 51). In one case, a metal detector had to be used to find the outlets. Even when the outlet headwall was visible and fairly clear, it was often necessary to use hand tools to clear dirt and debris out of the first foot or so of the outlet. Measuring Longitudinal Grade Because no coring was permitted within the test sections, coring was done in the transition sections just outside the test section limits. In some cases, it was evident which of the test section ends was higher than the outlet to be tested, but for those pavements with almost no longitudinal grade, it was necessary to determine which test section end was higher. The longitudinal grade was measured using a carpenter’s level with a digital display (Figure 52). Coring A core was cut through the pavement surface down to the top of the permeable base layer. The as-constructed layer thickness information (see Appendix A) was con- sulted to determine the depth of coring necessary to reach the top of the PATB layer. At some locations, it was not possible to remove all of the hot-mix asphalt (HMA) mate- rial above the permeable asphalt-treated base because of the HMA’s considerable thickness. At those locations, the cor- ing was advanced through the top of the permeable asphalt- treated base, so that water could flow down into the base during the testing. The coring was conducted by the state DOT (Figure 53) in all but one case; when one DOT was unable to provide a coring rig and operator, the consultant rented the necessary equipment and conducted the coring (Figure 54). Other Measurements Because coring had to be conducted outside the limits of the test section, but the locations of the drainage outlets inside the limits of the test sections were not consistent from site to site, it was necessary to obtain a variety of distance and elevation measurements that could be used to later calculate the length of the flow path. A measuring wheel was used to C H A P T E R 3 Field Testing of Drainage Systems

34 Figure 43. View from shoulder and close-up view of overgrown drainage outlet, Michigan SPS-2. Figure 44. View from shoulder and close up view of overgrown drainage outlet, Kansas SPS-2. Figure 45. View from shoulder and close-up view of overgrown drainage outlet, Iowa SPS-2.

35 Figure 46. View from above and close-up view of overgrown drainage outlet, Delaware SPS-2. Figure 47. Clearing drainage outlet, Texas SPS-1. Figure 48. Clearing drainage outlet, Texas SPS-1. measure the transverse distance from the core hole to the edge of the pavement, the longitudinal distance from the core hole to the drainage outlet, and the transverse distance from the edge of the pavement to the drainage outlet (Figure 55). A rod and a laser level were used to measure the elevation of the surface of the pavement next to the core hole, the top of the permeable asphalt-treated base layer in the core hole, the edge of the pavement at the core hole station, the edge of the pavement at the drainage outlet station, and the inside bot- tom edge of the drainage outlet pipe (Figure 56). The digital level was used to measure transverse and longitudinal slopes. Each core was photographed, and its thickness was measured (Figures 57 and 58). Measuring Inflow and Outflow The major pieces of equipment needed for the testing are shown in Figure 59. Water was run from a water truck pro- vided by the DOT, through a hose to a water pump, then through a flow meter (Figure 60), and finally into the core hole (Figures 61 and 62). The flow meter’s screen can display either

36 Figure 49. Drainage outlet before and after being uncovered, Florida SPS-1. Figure 50. Dual outlets, one cleared for testing, one packed with dirt and stone, Nevada SPS-1. Figure 51. Dual outlets, one cleared, one blocked by dirt and stone, Nevada SPS-2. Figure 52. Digital level for measuring transverse and longitudinal grades. the total volume of water used, in gallons, or the rate of water flow, in gallons per minute. The water pump was powered by a car battery. The water pump was not needed in those cases where the water head from the truck was sufficient to achieve maximum measurable flow through the flow meter or maxi- mum inflow capacity of the permeable asphalt-treated base. Normally the tests were conducted by first adjusting the flow rate to the maximum that the permeable asphalt-treated base could accommodate without water spilling out of the top of the core hole. The maximum inflow rate was recorded, and the flow rate was then reduced to a steady-state rate of 8 gal/min. If the maximum inflow capacity of the base was less than 8 gal/min, the inflow rate was set to a value that would maintain the water level in the core hole just below the pavement surface. Water was allowed to flow into the base until it was observed flowing out of the nearest downstream outlet (Figure 63); this usually took at least several minutes. Once free flow through

37 Figure 53. Coring by state DOT personnel, Arkansas SPS-1. Figure 54. Rented coring rig, Nevada SPS-1. the drainage system was established, a tracer dye was added to the inflowing water. In the pilot test, liquid soap had been used to attempt to measure the time to free flow (Figure 64), but a tracer dye was found to produce more clearly visible results (Figure 65). A stopwatch was used to measure the time to when outflow was first observed, the time to when tracer dye outflow was observed, and the time to when inflow was stopped. Patching Core Holes The core holes were patched by the research team, using similar materials, or were, at the state DOT’s discretion, left to be patched by the state’s own crews. Photos from the pilot tests at the Arkansas SPS-2 site pro- vide vivid evidence of a drain that functioned not only during the testing, but also prior to the testing. What appears to be soft mud at the end of the drainage outlet in Figure 66 is, in fact, a hardened buildup of residue from drainage flows that occurred before the pilot test. Figure 67 shows the damage to nearby vegetation that appears to be due to this outflow. The SPS-2 test sections at this site were not built directly on the subgrade or prepared fill, but rather on top of an old con- crete pavement that had been rubblized. Leaching of chemi- cals from this old rubblized concrete layer may be the cause of the vegetation damage seen in Figure 67. Drainage Flow Calculations The following general equation is used to determine the rate of flow through a porous medium: Q = k i A where Q = rate of flow through cross-sectional area (length/time), k = hydraulic conductivity of medium (length/time), i = hydraulic gradient (elevation head difference/length), and A = cross-sectional area (length2).

38 Figure 55. Measuring distances between core hole and drainage outlet. Figure 56. Rod and level measurement of elevation of bottom of drainage outlet pipe at headwall. Figure 57. Measuring thickness of AC lifts above PATB from core. Figure 58. Measuring thickness of PCC above PATB from core. Figure 59. Equipment for flow time testing.

39 Figure 60. Flow meter. Figure 61. Water truck, connecting hose, flow meter, car battery, and tracer dye. Figure 62. Water inflow from flow meter to PATB. Figure 63. Free flow is established when clear water flows out of drain.

This equation can be rearranged as follows to solve for the hydraulic conductivity, k, as a function of a known flow rate, hydraulic gradient, and area: k = Q / i A For this study, the variables in the above equation are defined as follows (and illustrated in Figures 68 and 69): k = estimated hydraulic conductivity of PATB (ft/day) Q = maximum inflow rate measured during field tests (measured in gal/min, converted to cu ft/day) i = hydraulic gradient measured in field (Δh / L) h = elevation head difference measured in field = (1 − 2 ) + 3 + 4 where 1 = elevation measured at pavement edge (ft), 2 = elevation measured at top of pavement at core hole (ft) 3 = pavement thickness above PATB (ft), and 4 = thickness of PATB (ft). L = flow length (ft)—the distance measured from core hole to pavement edge A = cross-sectional area of flow (sq ft)—the thickness of PATB (ft) × assumed width of flow plume through PATB (ft). The above equations are based on transverse flow (a longitudi- nal grade of 0%). As the longitudinal grade increases above 0%, both the hydraulic gradient and the flow length increase. The proportional increase is the same in both, however, making the above equation valid for any longitudinal gradient. An example of these calculations is provided below, using the measurements from the test at one of the drainage outlets at the Alabama SPS-1 site. Date: 08/18/03 SHRP site ID: 010107 Core hole test station: 0 - 68 GPS coordinates: N 32° 36.344' W 85° 15.027' Cross slope (%): 1.2 40 Figure 64. Bubbles from liquid soap used in pilot testing to attempt to measure time to free flow. Figure 65. Flow time measured from introduction to outflow of tracer dye.

Longitudinal grade (%): 0.8 Distance measures (ft) Core to edge: 5.7 Core to outlet: 80.0 Edge to outlet: 21.0 Elevation readings (ft) Top of pavement at core: 1.58 Top of PATB after coring: 1.96 Edge of pavement: 1.67 Edge at outlet: 2.50 Outlet: 5.92 Infiltration measures Steady-state infiltration rate (gal/min): 6 Time to first outflow (min:sec): 13:58 Cumulative inflow to tracer input (gal): 15 @ 2:33 Time to tracer outflow (min:sec): 13:58 Maximum inflow rate (gal/min): 6 41 Figure 66. Dual outlets, one flowing, and hardened buildup of past outflow residue, believed to be due to rubblizing of old concrete pavement under test sections, Arkansas SPS-2. Figure 67. Vegetation near drainage outlet appears damaged, possibly due to chemicals leached from old rubblized concrete layer, Arkansas SPS-2. ShoulderPavt PATB Base Metered Water Inflow, Q Phreatic Line Flow Length, L dh k = Q / iA = Q L / dh A A = Average Cross-Sectional Area of Flow Plume Figure 68. Illustration of parameters used in determining in-place base permeability.

Water inflow stopped (gal): 83 Cross slope (elevation measures) (%) 1.5 Longitudinal grade (elevation measures) (%) 1.0 Thickness of pavement above PATB (elevation measures) (ft): 0.38 Thickness of pavement above PATB (elevation measures) (in.): 4.5 Thickness of pavement above PATB (LTPP database) (ft): 0.38 Thickness of pavement above PATB (LTPP database) (in.): 4.6 Thickness of PATB (LTPP database) (ft) 0.30 Thickness of PATB (LTPP database) (in.) 3.6 Hpavt (ft): 0.38 Hpatb (ft): 0.30 Q (cu ft/day): 1,154 Δh (ft): 0.76 42 SHRP ID State Subgrade drainage class Drainage system permeability (ft/day) SPS-1 010100 Alabama Well drained 11,257 040100 Arizona Well drained No outflow 050100 Arkansas Somewhat poorly drained Not calculated 100100 Delaware Very poorly drained 11,245 120100 Florida Very poorly drained 10.056 190100 Iowa Well drained 10,140 200100 Kansas Well drained 5,712 220100 Louisiana Poorly drained 8,841 260100 Michigan Somewhat poorly drained 8,528 300100 Montana Well drained 11,437 310100 Nebraska Well drained Not tested 320100 Nevada – 6,058 350100 New Mexico – 17,545 390100 Ohio – No outflow 400100 Oklahoma Well drained 9,199 480100 Texas Well drained 6,224 510100 Virginia Well drained 7,987 550100 Wisconsin – – – 13,289 SPS-2 040200 Arizona Well to somewhat excessively drained 15,966 050200 Arkansas Moderately well drained Not calculated 060200 California 8,803 080200 Colorado Well to somewhat excessively drained 14,270 100200 Delaware Well drained 9,981 190200 Iowa 9,809 200200 Kansas Well drained 12,225 260200 Michigan Very poorly drained 10,581 320200 Nevada – 9,275 370200 North Carolina Well drained 15,291 380200 North Dakota Poorly drained 10,172 390200 Ohio – 9,539 530200 Washington Well drained 32,656 550200 Wisconsin – 15,697 MN/Road Minnesota Mix of well and poorly drained 11,239 Table 18. Summary of permeability calculations from field testing data. Corehole Flow Plume 3 ft Average Width (assumed) Pavement Edge Average Cross-Sectional Area of Flow Plume = 3 ft x PATB Thickness Figure 69. Illustration of plume of water from core hole to pavement edge.

L (ft): 5.7 i (ft/ft): 0.13 Assumed width of flow plume (ft): 3 A (sq ft): 0.9 k (ft/day): 9,583 Although the result obtained appears reasonable, since it falls within the expected range for a permeable asphalt-treated base, it should be noted that there is at least one limitation to this ap- proach to calculating the in-place permeability of the base. The actual value obtained for the permeability, k, is a function of the assumed width of the flow plume. In the above example, a flow plume width of 3 ft was assumed; for the purposes of compari- son with other test results obtained, a width of 3 ft was assumed for all of the outlets tested in this study. But there is really no way of knowing what the true flow plume width was for the partic- ular core hole test considered in the example, nor what it is for other tests at other locations. Had a width of 4 ft been assumed, the calculated k would have been reduced to 7,187 ft; on the other hand, had a width of 2.5 ft been assumed, the calculated k would have been 11,499 ft. The calculated k can be changed by thousands of feet per day simply by varying the value assumed for the width of the flow plume. This suggests that it is best not to place too much impor- tance on the actual values of the permeability values calculated using the procedure outlined above. They are more meaning- ful as indicators of the functioning of the subdrainage system. When no outflow occurs, on the other hand, this might or might not be due to a malfunctioning of the subdrainage sys- tem. If water fails to flow out of just one of several outlets at a site, for example, this suggests some localized problem in the system, such as a blockage in the longitudinal pipe. But if water fails to flow out of all of the outlets at a site, this sug- gests that the water introduced into the permeable base flowed downward into the subgrade. Field Testing Results The measurements obtained in the field testing and the permeability values calculated from these field measure- ments are shown in Appendix B (Tables B-1 through through B-17 for the SPS-1 sites, Tables B-18 through B-31 for the SPS-2 sites, and Table B-32 for the MnRoad site). The results of the field testing of the subdrainage systems at the SPS-1 and SPS-2 sites and the MnRoad site are summa- rized in Table 18. 43