Site Evaluations Cool Creek Site Information Site Location: Mt. Hood NF. 2.5 miles off US 26 on Still Ck Rd (FR 2612) Year Installed: 1984 Lat/Long: 121°53’11.02”W 45°17’53.01”N Watershed Area (mi2): 1.7 Stream Slope (ft/ft)1: 0.0519 Channel Type: Step-pool Bankfull Width (ft): 19.5 Survey Date: 4-6-2007 1 Water surface slope extending up to 20 channel widths up and downstream of crossing. Culvert Information Culvert Type: Open-bottom arch Culvert Material:Annular CMP Culvert Width: 13 ft Outlet Type: Projecting Culvert Length: 48 Inlet Type: Projecting Pipe Slope (structure slope): 0.022 Culvert Bed Slope: 0.015 (First hydraulic control upstream of inlet to first hydraulic control downstream of outlet.) Culvert width as a percentage of bankfull width: 0.66 Alignment Conditions: Culvert placement perpendicular to road likely differs from original steam channel alignment. Outlet possibly oriented further south than original alignment. Lateral scour pool at inlet may be associated with alignment. Energy is directed at left wall of pipe at inlet. Bed Conditions: Cobbles to small boulders in pipe. Similar to natural channel. Pipe Condition: Good condition. Only minor rust in places. Hydrology Discharge (cfs) for indicated recurrence interval 25% 2-yr Qbf2 2-year 21 70 83 5-year 10-year 50-year 100-year 134 172 262 302 Bankfull flow estimated by matching modeled water surface elevations to field-identified bankfull elevations. 2 Cool Creek A—35 Culvert Scour Assessment Points represent survey points Figure 1—Plan view map. A—36 Cool Creek Site Evaluations History The exact installation date is unknown, but the culvert was included in the 1987 Western Federal Lands Highway Division (WFLHD) “Oregon Culvert Fish Passage Survey.” The field survey for the WFLHD study was conducted on 10/19/1987. Special features that were noted included the observation that the footings were 4 feet below the streambed and “manmade pools were built at outlet with log and rock barriers.” With respect to fish passage, Oregon Department of Fish and Wildlife (ODFW) staff described the culvert as a “very good open bottom arch,” but they note that the “manmade pools at outlet may require periodic maintenance.” WFLHD staff rated overall culvert condition as “poor,” culvert capacity as “fair,” condition of foundation as “fair,” and with “negligible” outlet scour. The culvert hydraulics were considered “compatible” with the natural stream hydraulics. In their comments they note that “despite the poor installation procedures, the Cool Creek culvert is a good fish passage design.” The following are photos from the WFLHD study: Culvert inlet. Cool Creek Typical stream channel. It is unknown whether the log-drop structure 15 to 20 feet downstream of the outlet was placed during construction or was placed in response to scour of the culvert bed at some point following construction. The drop structure was present at the time of these 1987 photos. Site Description The Cool Creek culvert is a short bottomless arch that projects from the roadfill. At the inlet to the culvert the steep channel, predominately comprised of step-pool channel units, abruptly flattens. A deep pool has formed up against the left wall, creating a depositional bar along the right bank. The downstream half of the culvert consists of a relatively plane bed glide. Downstream of the culvert outlet, and extending to the confluence of Still Creek, there are a series of constructed log-drop structures that are serving as the hydraulic controls for this reach. Artificial log-drop structures are also present upstream of the culvert. The presence of artificial log drops upstream of the culvert required that the upstream representative reach be located a couple hundred yards upstream of the inlet. The upstream reach consisted of a moderate gradient steppool channel with active flood-plain terraces intermittent along both sides. Large material A—37 Culvert Scour Assessment through the reach provided the framework for the step/cascade units that made up the majority of the reach. A plunge pool adjacent to a log jam was the only pool within the reach and served as a reference cross section. downstream transition area was comparable to a representative segment in the upstream channel. There was no comparable segment for the upstream transition segment (H). See figure 2 and table 1. Survey Summary Scour Conditions Thirteen cross sections and a longitudinal profile were surveyed along Cool Creek in April 2007 to characterize the culvert and an upstream reference reach. No downstream reference reach was established due to the presence of artificial log drops. In the culvert, representative cross sections were taken through the pool and the glide. Three additional cross sections were surveyed downstream of the culvert to characterize the outlet as well as the expansion of flow. Two additional cross sections were surveyed upstream to characterize the inlet as well as the contraction of flow. Observed conditions Due to the complexity of the channel between the upstream representative reach and the culvert, two separate hydraulic models were run to evaluate the hydraulics through these reaches. While a longitudinal profile was surveyed, cross sections did not capture grade breaks and hydraulic controls between reaches. Four cross sections were surveyed to characterize the upstream representative reach; one at the upstream and downstream boundary, one along the top of a step and one through a pool. Profile Analysis Segment Summary The profile analysis resulted in 13 profile segments. The culvert consisted of one profile segment that extended into the culvert outlet transition area. The culvert segment was comparable to one representative segment in the upstream channel. A segment in the A—38 Footing scour—There was minor scour along the left bank footing at the inlet. Footings were not scoured to the base or undermined in any way that threatens structure integrity. Culvert bed adjustment—The culvert bed shows some flattening of the profile based on comparisons of the bed to the slope of the structure itself (assuming the culvert bed was constructed at the same gradient as the culvert structure). This flattening appears to be a combination of scour at the inlet and aggradation towards the outlet. A scour pool has formed along the left edge of the culvert, leaving a raised surface along the upper right channel within the culvert. Material has deposited through the downstream half of the culvert filling in the channel to resemble a glide. Backwatering and grade control has been provided by the presence of log-drop structures downstream of the culvert. Profile characteristics—The profile has a concave shape through the crossing (figure 2). This shape reflects scour at the inlet region, combined with aggradation in the downstream portion of the culvert and in the channel downstream of the culvert. A significant transition in the channel gradient occurs at the inlet to the culvert as the channel flows through a series of log-drop structures. It is this transition, along with the contraction of flow through the culvert that has created the scour along the left edge of the culvert. The low crown elevation of the culvert and the deep level of embedment suggest that the Cool Creek Site Evaluations installation of the Cool Creek crossing may have been placed too low in relation to the original channel profile, which may have occurred in order to avoid raising the road bed. Deep embedment during construction and/or deposition (partially related to the downstream drop structure) is contributing to reduced capacity during flood flows. Residual depths —The residual depth of the single pool within the culvert (segment G) is greater (0.94) than the upstream channel (0.360.82) (figure 21). Additionally, the downstream transition segment (F) has greater residual depths (0.43-0.54) than does the upstream channel (0.15-0.23). This suggests that the culvert has experienced additional scour beyond what is found in the channel outside of the crossing. No units of measure. Substrate—Substrate in the upstream channel is coarser than that found in the culvert, with more frequent large particles and less frequent smaller particle sizes (including sand and small gravels). A greater frequency of smaller particles in the culvert may be due to the gradient transition in the culvert and the reduced capacity at higher flows. Backwatering and reduced sediment transport capacity (i.e., low shear stress) allows for the deposition of smaller particles. All values are poorly sorted and positively skewed, which is typical for mountain streams (table 7). Although culvert sorting and skewness values fall outside the range of the natural channel, values do not diverge substantially. Pebble counts are provided at the end of the site summary. Predicted conditions Cross-section characteristics—Cross-sectional flow area, wetted perimeter, top width and the width-to-depth ratio are considerably reduced by the culvert (figure 12, figure 15, figure 17) reflecting the reduced capacity of the culvert. Hydraulic radius and maximum depth in the Cool Creek culvert (segment G) are similar to the upstream channel (segment L) for flows up to the Qbf and then become greater within the culvert above the Qbf (figure 14, figure 16). Flow area, top width, and maximum depth are similar for the downstream transition segment (F) as they are for the upstream channel (M) for the range of flows (figure 12, figure 15, figure 16). The wetted perimeter and hydraulic radius of the downstream transition (segment F) are similar to the upstream channel (segment M) up to the Q10, above which the wetted perimeter is less in the downstream transition than in the upstream channel, while the hydraulic radius is greater (figure 13, figure 14). The downstream transition segment (F) has a greater width-to-depth ratio as the upstream channel segment (M) for flows up to the Qbf, above which the transition segment’s width-todepth ratio is less than the upstream channel’s (figure 17). Shear stress—The shear stress within the culvert (segment G) is similar to that of the upstream channel (segment L) for all flows modeled (figure 10, figure 19). Similarly, the shear stress of the downstream transition segment (F) is similar to the upstream channel segment (M) for all flows modeled. However, the range of shear-stress values found in the transition segment is great and the median value is greater than the median value of the upstream channel. Excess shear—The excess shear analysis shows that the culvert and upstream channel have similar excess shear values for the 25-percent Q2 and the Qbf, but diverge above the Qbf (figure 20). It is above the Qbf that the potential for bed mobilization in the upstream channel increases at a greater rate than in the culvert. This corresponds with lower shear in the culvert at higher flows. Velocity—Velocity in the upstream culvert segment (G) is higher than the downstream representative segment (L) for all flows modeled (figure 11, figure 18). These higher culvert velocities correspond to the flow contraction A—39 Culvert Scour Assessment (reduction in flow area) caused by the culvert. Velocity at the downstream transition (F) is similar to the upstream representative segment (M) at all flows. Scour summary The Cool Creek culvert showed signs of bed adjustment including inlet scour and downstream aggradation. Scour at the inlet is likely related to inlet-control conditions (limited capacity) during floods. The downstream aggradation is likely caused by the grade control log-drop structures located downstream of the crossing that lower the local slope. The abrupt transition in gradient from the upstream log-drop structures to the culvert may be a function of the placement of the crossing. Continued aggradation in the downstream portion of the culvert could further reduce the capacity of the culvert. AOP Conditions Cross-section complexity—The sum of squared height differences in the culvert cross sections are not within the range of those in the channel cross sections (table 3). The upstream culvert cross section, taken through a pool has a higher value than the upstream channel. The inlet scour has created a deep pool along the left channel, leaving a higher bar surface along the right channel creating complexity in the channel. In contrast, the downstream culvert section, taken through the glide is relatively flat and featureless and represents the aggraded portion of the culvert which is backwatered by the downstream log-drop grade control structures. A—40 Profile complexity—Vertical sinuosity in the culvert is similar to that found in the upstream channel (segment L) (table 4). However, the vertical sinuosity of the downstream transition (segment F) is much higher than the upstream channel (segment M), a result of the grade control log structures. Depth distribution—There is more channel margin habitat in the culvert compared to the channel at 25-percent Q2 (table 5). Habitat units—The habitat-unit composition in the culvert is very different than the upstream channel (table 6). Whereas the culvert is made up of one long scour pool followed by a long glide, the upstream channel is composed of steep riffles interspersed by short pools, a step, and a plunge pool. Residual depths—The residual depth of the plunge pool in the culvert (0.95) is greater than that found in the upstream channel (0.36-0.82) (figure 21). Similarly, residual depths in the downstream transition (0.43-0.54) are greater than those in the upstream channel (0.15-0.23). Bed material—The bed material in the upstream channel is coarser than that found in the culvert. The distributions of the two pool units are fairly similar, with a slightly greater number of coarser particles and a noticeable lack of sand and very fine gravels in the upstream channel. The step has no rocks less than 11.3 millimeters and the glide has no rocks greater than 362 millimeters. The larger material in the natural channel may provide for velocity refuge for fish that exceeds what is available in the culvert. Cool Creek Site Evaluations Large woody debris—There was no LWD present in the culvert (table 8). The representative channel had high LWD abundance. LWD formed steps and scour pools in the channel outside the crossing and played a primary role in habitat-unit creation and complexity. Features in the culvert did not mimic the role of wood in the natural channel. The culvert, which has a low rise, is unlikely to be able to transport or retain LWD without considerable scour risk. Constructed wood-drop structures were present in the channel upstream and downstream of the crossing. AOP summary Some metrics, such as cross-section complexity and shallow-water habitat availability at 25-percent Q2 are greater in the culvert than the natural channel. However, cross-section shape (e.g., width-to-depth ratio and wetted perimeter) changes sharply at the crossing, with potential impacts on availability of channel margin habitat as flows increase. The habitat unit composition is very different in the culvert with only one long Cool Creek pool unit whose residual depth is greater than that found in the upstream channel. The greater residual depth may create good holding cover in the culvert; however, large substrate elements that are also important for velocity refuge are scarcer in the culvert. Design Considerations This culvert is deeply embedded and may have capacity limitations, especially if material continues to aggrade inside the structure. Because of the low height (rise) of the structure, woody debris may become easily impinged at the inlet or inside the culvert, further limiting capacity. Flattening of the gradient through downstream grade control has possibly kept material from scouring out over the years but it also could lead to further aggradation and capacity reduction that could cause overtopping during a large event. Increasing the culvert rise could ameliorate these potential issues, but it would require raising the road prism because of the low elevation of the road fill above the pipe. A—41 A—42 Relative Elevation (ft) 800 B C Log drop structures D 700 E XS 1 F XS 7 F M L G D o w n s tre a m T ra n s itio n R e p re s e n ta tiv e C h a n n e l S egment C u lv e rt S e g m e n t Table 1—Segment comparisons XS 6 H XS 5: Pebble count (pool) 600 XS 4: Pebble count (glide) XS 2 XS 3 C ulvert G Figure 2—Cool Creek longitudinal profile. 480 490 A 500 Relative elevation (feet) 510 520 530 540 I XS 10 XS 11 400 D is ta nc e a long bed (feet) J 35% 34% % D iffe re n c e in G ra d ie n t 500 XS 8 XS 9 Log drop structures 300 S egm ent A B C D E F G H I J K L M K S egm ent L ength (ft) 22 17 47 12 36 26 76 22 72 66 245 78 68 200 XS 12 L S egm ent G ra dient 0.046 0.075 0.022 0.130 0.030 0.060 0.015 0.101 0.054 0.080 0.060 0.023 0.039 100 Representative Channel XS 13: Pebble count (pool) XS 15 XS 14: Pebble count (step) M 0 Culvert Scour Assessment Cool Creek Cool Creek G L M 0.1136—0.1164 0.1095—0.1179 0.058—0.123 0.123—0.1279 2 2 4 2 # of measured XSs 6 9 6 3 # of interpolated XSs 1 Obtained using equation from Jarrett (1984): n = 0.39S0.38R-0.16, where S=stream slope; R=hydraulic radius. Jarrett’s equation only applied within the following ranges: S = 0.002 to 0.08, R = 0.5 ft to 7 ft. For cross sections outside these ranges, n was computed either from adjacent sections that fell within the ranges, using the guidance of Arcement and Schneider (1987), or from the HEC-RAS recommendations for culvert modeling. F SegmentRange of Manning’s n values1 Table 2—Summary of segments used for comparisons Site Evaluations A—43 ) t f ( n o i t a v e l E 496 0 1 526 528 530 532 534 536 0 * 5 . 1 . . . . 1 1 2 2 1 * 5 . 2 20 . . . . 2 1 Figure 3—HEC-RAS profile. Elevation (ft) 498 500 502 504 506 508 510 512 Elevation (ft) ) t f ( A—44 n o i t a v e l E 3 4 . . . . 2 1 40 3 . 4 . . . . 2 1 * 5 6 . 4 5 . . . . . . 3 3 1 1 60 80 * 3 3 3 8 . 3 1 5 . 5 6 100 * 5 . 6 Main Channel Dis tance (ft) 5 2 . 5 100 1 . 7 7 * 5 2 . 7 b) Culvert Reach Main Channel Distance (ft) 5 . 3 1 * 6 6 6 6 . 3 1 4 1 * 4 . 7 5 2 . 4 1 * 5 5 . 7 150 5 . 4 1 120 7 . 7 * 5 8 . 7 5 7 . 4 1 8 * 5 4 . 8 5 1 140 Stations with decimal values are interpolated cross sections placed along the surveyed profile. 50 . . . . 2 1 . . . * 3 . 3 1 a) Upstream Channel Reach 9 . 8 9 200 160 Le gen d Ground W S 20 c fs W S Qbf W S Q10 W S Q50 W S Q100 L e ge nd G rou nd WS 20 cfs WS Q bf WS Q 10 WS Q 50 WS Q 100 Culvert Scour Assessment Cool Creek Elevation (ft) n o i t a v e l E ) t f ( n o i t a v e l E ) t f ( Cool Creek 526 0 20 .1095 20 RS = 13 40 Station (ft) 60 60 Station (ft) .1095 40 80 .1095 80 .1152 100 Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd 100 B a n k S ta L e ve e G ro u n d W S 2 0 c fs W S Qb f W S Q1 0 W S Q5 0 W S Q1 0 0 L e ge n d n o i t a v e l E ) t f ( Elevation (ft) 526 528 530 532 534 536 530 532 534 536 538 0 0 20 20 .1162 .1136 40 60 60 .1162 RS = 12 Station (ft) Station (ft) 40 .1136 RS = 14 .1136 80 .1162 80 W S Qb f W S Q1 0 W S Q5 0 Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd B a n k S ta L e ve e G ro u n d W S 2 0 c fs 100 100 L e ge n d W S Q1 0 0 Figure 4­—Cross-section plots. Only measured cross sections are included. Manning’s n values are included at the top of the cross section. The stationing (RS) corresponds to the stationing on the HEC-RAS profile. Green arrows define the ineffective flow areas. Black arrows represent points identified in the field as the bankfull channel boundary. Only those points identified in the field and supported by hydraulic and topographic analyses are shown below. Elevation (ft) 528 530 0 .1152 n o i t a v e l E Elevation (ft) 532 534 536 532 534 536 538 540 .1152 RS = 15 ) t f ( Site Evaluations A—45 ) t f ( A—46 Elevation (ft) n o i t a v e l E ) t f ( n o i t a v e l E 502 -20 -10 0 10 20 Station (ft) 30 . 1 3 9 4 RS = 7 Station (ft) .1394 0 40 20 50 60 40 70 80 60 Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd n o i t a v e l E ) t f ( Elevation (ft) 500 -10 502 504 506 508 510 -40 506 508 510 512 514 0 10 -20 20 .075 30 0 Station (ft) 40 .0653 RS = 6 50 20 .137 Station (ft) . 1 3 7 RS = 8 60 .075 . 1 3 7 70 40 80 60 90 B a n k S ta In e ff G ro u n d W S 2 0 c fs W S Qb f W S Q1 0 W S Q5 0 W S Q1 0 0 L e ge n d Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd Figure 4­—Cross-section plots. Only measured cross sections are included. Manning’s n values are included at the top of the cross section. The stationing (RS) corresponds to the stationing on the HEC-RAS profile. Green arrows define the ineffective flow areas. Black arrows represent points identified in the field as the bankfull channel boundary. Only those points identified in the field and supported by hydraulic and topographic analyses are shown below. (continued) Elevation (ft) 504 506 .1394 -20 .137 n o i t a v e l E Elevation (ft) 508 510 512 -40 508 510 512 514 516 . 1 3 7 RS = 9 ) t f ( Culvert Scour Assessment Cool Creek Cool Creek Elevation (ft) n o i t a v e l E ) t f ( n o i t a v e l E ) t f ( 0 -10 20 .1195 10 -20 10 50 20 Station (ft) 40 .1195 RS = 3 Station (ft) 30 0 60 .1195 30 70 40 80 50 60 90 B a n k S ta In e ff G ro u n d W S 2 0 c fs W S Qb f W S Q1 0 W S Q5 0 W S Q1 0 0 L e ge n d Bank St a n o i t a v e l E ) t f ( Elevation (ft) -20 500 502 504 506 508 500 -40 502 504 506 508 510 -10 -30 0 -20 10 .1264 -10 10 50 30 .1264 40 20 Station (ft) 30 .1264 RS = 2 Station (ft) 20 0 .0656 RS = 4 Note: n values for first profile. 60 40 50 70 60 80 B a n k S ta In e ff G ro u n d W S 2 0 c fs W S Qb f W S Q1 0 W S Q5 0 W S Q1 0 0 L e ge n d Bank St a Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd Figure 4­—Cross-section plots. Only measured cross sections are included. Manning’s n values are included at the top of the cross section. The stationing (RS) corresponds to the stationing on the HEC-RAS profile. Green arrows define the ineffective flow areas. Black arrows represent points identified in the field as the bankfull channel boundary. Only those points identified in the field and supported by hydraulic and topographic analyses are shown below. (continued) Elevation (ft) 500 -10 502 504 -30 Ground WS 20 c fs WS Qbf WS Q10 WS Q50 WS Q100 L e ge nd n o i t a v e l E Elevation (ft) 506 508 510 -40 500 502 504 506 508 .0629 RS = 5 Note: n values for first profile. ) t f ( Site Evaluations A—47 Elevation (ft) A—48 -20 498 500 502 504 506 -10 0 10 .1279 20 40 .1279 Station (ft) 30 .1279 RS = 1 50 60 70 80 B a n k S ta In e ff G ro u n d W S 2 0 c fs W S Qb f W S Q1 0 W S Q5 0 W S Q1 0 0 L e ge n d Figure 4­—Cross-section plots. Only measured cross sections are included. Manning’s n values are included at the top of the cross section. The stationing (RS) corresponds to the stationing on the HEC-RAS profile. Green arrows define the ineffective flow areas. Black arrows represent points identified in the field as the bankfull channel boundary. Only those points identified in the field and supported by hydraulic and topographic analyses are shown below. (continued) n o i t a v e l E ) t f ( Culvert Scour Assessment Cool Creek Cool Creek Flow Area (sq ft) Flow Area (sq ft) 0 20 40 60 80 100 0 20 40 60 80 100 120 140 0 0 F 20 Figure 5—Flow Area (total) profile plot. w o l F a e r A q s ( ) t f w o l F a e r A q s ( ) t f L 50 40 G 80 100 Main Channel Distance (ft) 100 H I 150 Flow 120 b) Culvert Reach (sections 9-1) Main Channel Distance (ft) Culvert 60 M a) Upstream Reach (sections 15-12) 140 Flow 200 160 Flow Area 20 cfs Flow Area Qbf Flow Area Q10 Flow Area Q50 Flow Area Q100 Legend Flow Area 20 cfs Flow Area Qbf Flow Area Q10 Flow Area Q50 Flow Area Q100 Legend Site Evaluations A—49 A—50 W.P. Total (ft) W.P. Total (ft) 5 10 15 20 25 30 35 40 10 20 30 40 50 60 70 80 0 0 F Figure 6—Wetted perimeter. . P . W l a t o T ) t f ( . P . W l a t o T ) t f ( 20 L 50 40 G 80 100 Main Channel Distance (ft) 100 H b) Culvert Reach (sections 9-1) Main Channel Distance (ft) Culvert 60 I 150 Flow 120 M a) Upstream Reach (sections 15-12) 140 Flow 200 160 W.P. Total 20 cfs W.P. Total Qbf W.P. Total Q10 W.P. Total Q50 W.P. Total Q100 Legend W.P. Total 20 cfs W.P. Total Qbf W.P. Total Q10 W.P. Total Q50 W.P. Total Q100 Legend Culvert Scour Assessment Cool Creek Cool Creek r d y H s u i d a R ) t f ( 1.6 1.8 2.0 2.2 0 Hydr Radius (ft) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.4 0.6 0.8 1.0 1.2 1.4 Hydr Radius (ft) 0 F 20 Figure 7—Hydraulic Radius. r d y H s u i d a R ) t f ( L 50 40 G 80 100 120 Main Channel Distance (ft) 100 H I Flow 150 b) Culvert Reach (sections 9-1) Main Channel Distance (ft) Culvert 60 M a) Upstream Reach (sections 15-12) 140 Flow 200 160 Hydr Radius 20 cfs Hydr Radius Qbf Hydr Radius Q10 Hydr Radius Q50 Hydr Radius Q100 Legend Hydr Radius 20 cfs Hydr Radius Qbf Hydr Radius Q10 Hydr Radius Q100 Hydr Radius Q50 Legend Site Evaluations A—51 A—52 Top Width (ft) 5 10 15 20 25 30 35 10 20 30 40 50 60 70 0 0 Figure 8—Top Width. p o T Top Width (ft) h t d i W ) t f ( p o T h t d i W ) t f ( F 20 L 50 40 G 80 100 Main Channel Distance (ft) 100 H b) Culvert Reach (sections 9-1) Main Channel Distance (ft) Culvert 60 I 150 Flow 120 M a) Upstream Reach (sections 15-12) 140 Flow 200 160 Top Width 20 cfs Top Width Qbf Top Width Q10 Top Width Q50 Top Width Q100 Legend Top Width 20 cfs Top Width Qbf Top Width Q10 Top Width Q50 Top Width Q100 Legend Culvert Scour Assessment Cool Creek Cool Creek x a M Max Chl Dpth (ft) l h C h t p D ) t f ( x a M l h C h t p D ) t f ( 0 0 0 F 20 Figure 9—Maximum Depth. Max Chl Dpth (ft) 1 2 3 4 5 6 0 1 2 3 4 5 6 L 50 40 Culvert G 60 100 120 Main Channel Distance (ft) 100 H 150 b) Culvert Reach (sections 9-1) Main Channel Distance (ft) 80 I Flow M a) Upstream Reach (sections 15-12) 140 Flow 200 160 Max Chl Dpth 20 cfs Max Chl Dpth Qbf Max Chl Dpth Q10 Max Chl Dpth Q50 Max Chl Dpth Q100 Legend Max Chl Dpth 20 cfs Max Chl Dpth Qbf Max Chl Dpth Q10 Max Chl Dpth Q50 Max Chl Dpth Q100 Legend Site Evaluations A—53 A—54 r a e h S 12 14 12 Shear Chan (lb/sq ft) F 20 L 50 40 Figure 10—Shear stress (channel) profile. 0 2 4 6 8 10 0 0 14 0 2 4 6 8 10 Shear Chan (lb/sq ft) n a h C q s / b l ( ) t f r a e h S n a h C q s / b l ( ) t f G 80 100 Main Channel Distance (ft) 100 H I 150 Flow 120 b) Culvert Reach (sections 9-1) Main Channel Distance (ft) Culvert 60 M a) Upstream Reach (sections 15-12) 140 Flow 200 160 Shear Chan 20 cfs Shear Chan Qbf Shear Chan Q10 Shear Chan Q50 Shear Chan Q100 Legend Shear Chan 20 cfs Shear Chan Qbf Shear Chan Q10 Shear Chan Q50 Shear Chan Q100 Legend Culvert Scour Assessment Cool Creek Cool Creek Vel Chnl (ft/s) Vel Chnl (ft/s) 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 0 0 F 20 L 50 40 Figure 11—Velocity (channel) profile plot. l e V l n h C ) s / t f ( l e V l n h C ) s / t f ( G Main Channel Distance (ft) 80 100 Main Channel Distance (ft) 100 H b) Culvert Reach (sections 9-1) Culvert 60 Flow I 150 Flow 120 M a) Upstream Reach (sections 15-12) 140 200 160 Vel Chnl 20 cfs Vel Chnl Qbf Vel Chnl Q10 Vel Chnl Q50 Vel Chnl Q100 Legend Vel Chnl 20 cfs Vel Chnl Qbf Vel Chnl Q10 Vel Chnl Q50 Vel Chnl Q100 Legend Site Evaluations A—55 A—56 80 120 0 F (ds trans) 40Area (ft2) Flow Flow Area (ft2) L 25%Q2 G (culv) Figure 12—Flow Area (total). 50 M F (ds trans) G (culv) Qbf L 100% of the values M F (ds trans) 50% of the values G (culv) 10 L M F (ds trans) M inim um value th G (culv) 25 th percentile Median (aka 50 75 th percentile Maxim um value Box Plot Explanation 50 L percentile) M F (ds trans) G (culv) 100 L M Culvert Scour Assessment Cool Creek Cool Creek Wetted Perimeter (ft) F (ds trans) L 25%Q2 G (culv) F (ds trans) L 25%Q2 G (culv) Figure 14—Hydraulic Radius. Hydraulic Radius (ft) 0 1 Hydraulic Radius (ft) 2 3 4 Figure 13—Wetted Perimeter. 0 20 Perimeter (ft) Wetted 40 60 80 M M F (ds trans) F (ds trans) G (culv) G (culv) Qbf Qbf L L M M F (ds trans) F (ds trans) G (culv) G (culv) 10 10 L L M M F (ds trans) F (ds trans) G (culv) G (culv) 50 50 L L M M F (ds trans) F (ds trans) 100 100 G (culv) G (culv) L L M M Site Evaluations A—57 A—58 Top Width (ft) F (ds trans) F (ds trans) L 25%Q2 G (culv) Figure 16—Maximum Depth. Maximum Depth (ft) 0 Maximum Depth (ft) 2 4 6 L 25%Q2 G (culv) Figure 15—Top Width. 0 Top Width (ft) 20 40 60 80 M M F (ds trans) F (ds trans) Qbf Qbf G (culv) G (culv) L L M M F (ds trans) F (ds trans) G (culv) G (culv) 10 10 L L M M F (ds trans) F (ds trans) 50 50 G (culv) G (culv) L L M M F (ds trans) F (ds trans) 100 100 G (culv) G (culv) L L M M Culvert Scour Assessment Cool Creek Cool Creek Width-to-depth Ratio F (ds trans) L 25%Q2 G (culv) M Velocity (ft/sec) F (ds trans) L 25%Q2 G (culv) Figure 18—Velocity (channel). 0 Velocity (ft/sec) 2 4 6 8 M Figure 17—Width-to-depth Ratio. 0 10 Width-to-depth Ratio 20 30 40 F (ds trans) F (ds trans) G (culv) Qbf Qbf G (culv) L L M M F (ds trans) F (ds trans) G (culv) G (culv) 10 10 L L M M F (ds trans) F (ds trans) G (culv) G (culv) 50 50 L L M M F (ds trans) F (ds trans) 100 100 G (culv) G (culv) L L M M Site Evaluations A—59 A—60 Shear Stress (lbs/ft2) F (ds trans) L 25%Q2 G (culv) M Excess Shear (Applied/tcrit) 150 200 Dis charge (cfs ) F (ds trans) G (culv) 250 Qbf L 300 M 350 F (ds trans) Figure 20—Excess Shear Stress. Excess shear stress is the channel shear divided by the critical shear for bed entrainment of the D84 particle size. Values of excess shear greater than 1 indicate bed movement for the D84 particle size. Excess Shear (Applied / tcrit) 0 0 50 100 0.5 1 1.5 2 2.5 3 Figure 19—Shear Stress (channel). 0 4 Shear Stress (lbs/ft2) 8 12 16 10 L M F (ds trans) Culvert (DS pebble count) - glide US Channel (US pebble count) - step G (culv) G (culv) 50 L M F (ds trans) G (culv) 100 L M Culvert Scour Assessment Cool Creek Site Evaluations Table 3—Sum of squared height difference Reach XS LocationUnit type Culvert Sum of squared height difference Within range of channel conditions? US Pool 0.10 No DS Glide 0.03 No US Step 0.09 DS Pool 0.07 Upstream Table 4—Vertical sinuosity Segment A DS Channel 1.040 B DS Channel 1.046 C DS Channel 1.049 D DS Channel 1.001 E DS Channel 1.018 F DS transition 1.018 G Culvert 1.002 H US transition 1.006 I US Channel 1.015 J US Channel 1.015 K US Channel 1.007 L US Channel 1.003 M US Channel 1.002 Cool Creek LocationVertical Sinuosity (ft/ft) A—61 Culvert Scour Assessment Table 5—Depth distribution Reach XS Location 25% Q2 Culvert Within range of channel conditions? US 2 No DS 0 No US 4 DS 4 Upstream Table 6—Habitat unit composition Percent of surface area Reach Pool Glide Riffle Step Culvert 32% 44% 0% 0% Upstream Channel 15% 0% 76% 7% 3.5 2.5 2.0 1.5 Segment M Segment L Segment I Segment H: US Transition Segment G: Culvert Segment F: DS Transition Segment B Segment A Segment E 0.0 Segment K Segment D 0.5 Segment J Residual depth (ft) 1.0 Segment C Residual depth (ft) 3.0 Culvert SegmentISegment Segment J Segment L Segment A Segment B Segment C Segment D Segment E K Segment M Segment G: Segment F: Segment H: DS Transition US Transition Figure 21—Residual depths. A—62 Cool Creek Site Evaluations Table 7—Bed material sorting and skewness Reach XS LocationUnit Type Sorting Within range of channel conditions? Skewness Within range of channel conditions? Culvert US Pool 1.34 No 0.16 No DS Glide 1.70 No 0.15 No Upstream US Step 1.45 0.23 DS Pool 1.64 0.26 Table 8—Large woody debris Reach Pieces/Channel Width Culvert 0 Upstream 2.6 Terminology: US = Upstream DS = Downstream RR = Reference reach XS = Cross section Cool Creek A—63 Culvert Scour Assessment View upstream through culvert. View downstream through culvert. Culvert—downstream pebble count, glide. Culvert—upstream pebble count, pool. Upstream reference reach—upstream pebble count, riffle. Upstream reference reach—downstream pebble count, pool. A—64 Cool Creek Site Evaluations View upstream from culvert. Cool Creek View downstream from culvert (log step in center). A—65 Culvert Scour Assessment Cross section: Upstream Reference Reach—Upstream Pebble Count Material sand very fine gravel fine gravel fine gravel medium gravel medium gravel coarse gravel coarse gravel very coarse gravel very coarse gravel small cobble medium cobble large cobble very large cobble small boulder sm all boulder medium boulder large boulder very large boulder bedrock S ize C las s (mm) <2 2-4 4 - 5.7 5.7 - 8 8 - 11.3 11.3 - 16 16 - 22.6 22.6 - 32 32 - 45 45 - 64 64 - 90 90 - 128 128 - 180 180 - 256 256 - 362 362 - 512 512 - 1024 1024 - 2048 2048 - 4096 Bedrock C ount Item % C umulative % 0 0 0 0 0 4 0 6 7 8 18 4 18 15 7 9 3 0 0 0 0% 0% 0% 0% 0% 4% 0% 6% 7% 8% 18% 4% 18% 15% 7% 9% 3% 0% 0% 0% 0% 0% 0% 0% 0% 4% 4% 10% 17% 25% 43% 47% 66% 81% 88% 97% 100% 100% 100% 100% 20 100% 18 Frequency 14 60% 12 10 40% 8 Frequency 20% 6 4 Cumulative Frequency 80% 16 Cumulative Frequency 0% 2 Bedrock 2048-4096 512-1024 1024-2048 362-512 256-362 180-256 128-180 64-90 90-128 45-64 32-45 22.6-32 16-22.6 8-11.3 11.3-16 2-4 4 - 5.7 - 8 32 -45 45-64 64- 90 8 - 11.3 90 - 128 Bedrock 11.3 16- 16 -22.6 22.6- 32 128180 - 180 256 - 256 362 - 362 - 512 512 - 1024 1024 2048 2048 - 4096 Particle Size Category (mm) Partic le Siz e Category (mm) S ize C las s S ize percent finer than (mm) D5 D16 D50 D84 D95 D100 25 41 130 293 450 650 A—66 5.7-8 <2 4-5.7 <2 -20% 2-4 0 Sorting Coefficient: Material P ercent C ompos ition Sand Gravel Cobble Boulder Bedrock 0% 25% 56% 19% 0% 1.45 Skewness Coefficient:0.23 Cool Creek Site Evaluations Cross section: Upstream Reference Reach—Downstream Pebble Count Material S ize C las s (mm) sand very fine gravel fine gravel fine gravel medium gravel medium gravel coarse gravel coarse gravel very coarse gravel very coarse gravel small cobble medium cobble large cobble very large cobble small boulder small boulder medium boulder large boulder very large boulder bedrock <2 2-4 4 - 5.7 5.7 - 8 8 - 11.3 11.3 - 16 16 - 22.6 22.6 - 32 32 - 45 45 - 64 64 - 90 90 - 128 128 - 180 180 - 256 256 - 362 362 - 512 512 - 1024 1024 - 2048 2048 - 4096 Bedrock C ount Item % C umulative % 0 0 3 3 2 5 5 4 5 13 15 9 16 10 4 6 0 0 0 0 0% 0% 3% 3% 2% 5% 5% 4% 5% 13% 15% 9% 16% 10% 4% 6% 0% 0% 0% 0% 0% 0% 3% 6% 8% 13% 18% 22% 27% 40% 55% 64% 80% 90% 94% 100% 100% 100% 100% 100% 18 100% 16 80% Cumulative Frequency Frequency 14 12 60% 10 40% 8 Frequency 6 20% 4 2 Bedrock 2048-4096 1024-2048 362-512 512-1024 256-362 180-256 128-180 64-90 90-128 45-64 32-45 22.6-32 16-22.6 11.3-16 5.7-8 8-11.3 <2 4-5.7 <2 -20% 2-4 0 2-4 4 - 5.7 - 8 32 -45 45-64 64- 90 8 - 11.3 90 - 128 Bedrock 11.3 16- 16 -22.6 22.6- 32 128180 - 180 256 - 256 362 - 362 - 512 512 - 1024 1024 2048 2048 - 4096 Particle Size Category (mm) Partic le Siz e Category (mm) S ize C las s S ize percent finer than (mm) D5 D16 D50 D84 D95 D100 8 22 80 220 410 490 Cool Creek Cumulative Frequency 0% Sorting Coefficient: Material P ercent C ompos ition Sand Gravel Cobble Boulder Bedrock 0% 40% 50% 10% 0% 1.64 Skewness Coefficient:0.26 A—67 Culvert Scour Assessment Cross section: Culvert—Upstream Pebble Count Ma terial C ount Item % C umula tive % sand very fine gravel fine gravel fine gravel medium gravel medium gravel coarse gravel <2 2-4 4 - 5.7 5.7 - 8 8 - 11.3 11.3 - 16 16 - 22.6 S ize R ange (mm) 1 3 2 4 0 5 2 1% 3% 2% 4% 0% 5% 2% 1% 4% 6% 10% 10% 15% 18% coarse gravel very coarse gravel very coarse gravel small cobble medium cobble large cobble very large cobble small boulder 22.6 - 32 32 - 45 45 - 64 64 - 90 90 - 128 128 - 180 180 - 256 256 - 362 10 10 9 15 12 9 11 2 10% 10% 9% 15% 12% 9% 11% 2% 28% 38% 47% 63% 75% 85% 96% 98% small boulder medium boulder large boulder very large boulder bedrock 362 - 512 512 - 1024 1024 - 2048 2048 - 4096 > 4096 2 0 0 0 0 2% 0% 0% 0% 0% 100% 100% 100% 100% 100% 16 100% 14 90% Frequency 70% 10 60% 8 50% 40% 6 Frequency 30% 4 20% 2 10% Cumulative Frequency >4096 2048-4096 512-1024 1024-2048 362-512 256-362 180-256 128-180 64-90 90-128 45-64 32-45 22.6-32 16-22.6 8-11.3 11.3-16 5.7-8 <2 4-5.7 <2 0% 2-4 0 Cumulative Frequency 80% 12 2-4 4 - 5.7 5.7 - 8 > 4096 32 -45 45-64 64- 90 8 - 11.3 90 - 128 11.316 - 16 -22.6 22.6- 32 128180 - 180 256 - 256 362 - 362 - 512 512 - 1024 Particle Size Category (mm) 1024 2048 - 2048 - 4096 Particle Size Category (mm) S ize C las s S ize perc ent finer than (mm) D5 D16 D50 5 19 70 D84 D95 D100 176 242 490 A—68 Sorting Coefficient: Materia l P erc ent C ompos ition Sand Gravel Cobble Boulder Bedrock 1% 46% 48% 4% 0% 1.34 Skewness Coefficient:0.16 Cool Creek Site Evaluations Cross section: Culvert –Downstream Pebble Count Material S ize C la s s (mm) C ount Item % C um ulative % 0 0% 0% very fine gravel 2 - 4 fine gravel 4 - 5.7 3 1 3% 1% 3% 4% fine gravel 5.7 - 8 medium gravel 8 - 11.3 2 3 2% 3% 6% 9% medium gravel 11.3 - 16 coarse gravel 16 - 22.6 coarse gravel 22.6 - 32 5 4 5 5% 4% 5% 14% 18% 23% 17 16 17% 16% 39% 55% small cobble 64 - 90 medium cobble 90 - 128 17 8 17% 8% 72% 79% large cobble 128 - 180 very large cobble 180 - 256 15 4 15% 4% 94% 98% small boulder 256 - 362 small boulder 362 - 512 2 0 2% 0% 100% 100% 0 0 0 0% 0% 0% 100% 100% 100% 0 0% 100% sand <2 very coarse gravel 32 - 45 very coarse gravel 45 - 64 medium boulder 512 - 1024 large boulder 1024 - 2048 very large boulder 2048 - 4096 bedrock Bedrock 18 16 90% 14 80% 70% Frequency 12 60% 10 50% 8 Frequency 6 40% 30% 4 20% 2 10% >4096 2048-4096 512-1024 1024-2048 362-512 256-362 180-256 128-180 64-90 90-128 45-64 32-45 22.6-32 16-22.6 8-11.3 11.3-16 5.7-8 4-5.7 <2 <2 2-4 4 - 5.7 - 8 32 -45 45-64 64- 90 8 - 11.3 90 - 128 Bedrock 11.3 16- 16 -22.6 22.6- 32 180 - 180 256 - 256 362 - 362 - 512 Particle Size Category128 (mm) 512 - 1024 1024 2048 - 2048 - 4096 Partic le Siz e Category (mm) S ize C las s S ize perc ent finer than (mm) D5 D16 D50 D84 D95 D100 6 20 60 140 190 300 Cool Creek Cumulative Frequency 0% 2-4 0 Cumulative Frequency 100% Sorting Coefficient: Material 1.70 P erc ent C ompos ition Sand Gravel 0% 55% Cobble Boulder Bedrock 43% 2% 0% Skewness Coefficient:0.15 A—69