Loading...
HomeMy WebLinkAbout750391.tiff PRELIMINARY ENGINEERING REPORT DRAINAGE STUDY FIRESTONE-FREDERICK AREA WELD COUNTY, COLORADO Project No. 75-1 -CIV-0253 September, 1975 NELSON, HALEY, PATTERSON and QUIRK, INC. 2021 Clubhouse Drive Greeley, Colorado 750391 ENGINEERING ►-Zt NELSON, HALEY, PATTERSON and QUIRK, INC. CONSULTANTS 2021 CLUBHOUSE DRIVE GREELEY, COLORADO 80631 303: 356-4444 October 14, 1975 _ Board of Weld County Commissioners Post Office Box 758 Greeley, Colorado 80631 Gentlemen: We are pleased to submit our report concerning the storm drainage improvements in the Firestone-Frederick area of the county. The objective of the report is to identify the major problems in the area and propose conceptional and preliminary solutions to these problems. The county's desire to minimize capital expenditures as much as possible was taken into consideration for all of the proposed alternates. In submitting this report, we are pleased to acknowledge the cooperation given us in our work by the County Engineer, Mr. Richard Straub, upon whom we have had to call on for information. Finally, we wish to assure the county that we will be pleased to further our serivce in any way the county may desire. Sincerely yours, NELSON, HALEY, PATTERSON 1A and QUIRK, INC. �`ta (( � i.0 Brian J. Jas Yot� P. E. Project Manager BJJ/bb OFFICES IN GREELEY, DENVER, COLORADO SPRINGS, GRAND JUNCTION, COLORADO; RIVERTON, WYOMING TABLE OF CONTENTS Page No. SECTION I - INTRODUCTION Authorization 1 Purpose of Study 1 Scope of Work 1 SECTION II - STUDY AREA Topography and Natural Drainage 3 Subbasins 3 SECTION III - HYDROLOGY Storm Frequency 5 Hydrograph Procedure 5 Runoff Analysis 9 SECTION IV - HYDRAULIC ANALYSIS - EXISTING FACILITIES Main Channel 10 Other Facilities 13 SECTION V - PROPOSED FACILITIES - MAIN CHANNEL General 17 Alternate No. 1 17 Alternate No. 2 21 _ Advantages and Disadvantages 23 Other Alternatives 25 SECTION VI - PROPOSED FACILITIES - TRIBUTARY AREAS General 27 Subbasin E10 27 Subbasins D�3 and 024 28 $ubbasins D21 , D22, D11 , D12 29 Subbasins C10, C11 , and C12 31 Subbasins B10, 811 , and 812 31 Subbasins A10, All , and Al2 31 SECTION VIII - COST ESTIMATES General 34 Unit Costs 34 Cost Summaries Main Channel - Alternate No. 1 36 Main Channel - Alternate No. 2 37 Tributary Areas 38 Page No. SECTION VIII - SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary and Conclusions 40 Recommendations 42 SECTION I INTRODUCTION AUTHORIZATION This study of drainage problems in the Firestone-Frederick area of Weld County, Colorado is authorized under the terms of a contract dated April 2, 1975, between Nelson, Haley, Patterson and Quirk, Inc. , Greeley, Colorado and the Board of County Commissioners , Weld County, Colorado. PURPOSE OF STUDY The towns of Firestone, Evanston, and Frederick are presently drained by an open channel that skirts their western boundaries . The channel varies from a well defined channel of moderate hydraulic capacity in the Frederick area to a poorly defined channel of limited hydraulic capacity in the Evanston- - Firestone area. This lack of an adequate channel has resulted in numerous instances of flooding. In addition to the main channel problems, other drainage problems exist in the channel ' s tributary areas. These problems have resulted from urbanization of relatively flat land without consideration for proper drainage courses or facilities. The purpose of this study is to evaluate the storm drainage problems and to make recommendations for the repair, modification, replacement and/or con- struction of drainage facilities in the area. SCOPE OF WORK The scope of work for this study was divided into four phases : Phase I consisted of the preparation of topographic mapping of the Firestone-Frederick area at a scale of one inch equal to one hundred feet. -1 - Phase II consisted of a review of the drainage area including the development of such hydrologic factors such as rainfall design frequencies, runoff coefficients, infiltration allowances, and other pertinent data. Preliminary runoff computations were made to determine present flooding conditions. In addition, the Sullivan and Stanley irrigation ditches were analyzed for ditch overflow and the possible use as a means of run- off disposal . Phase III consisted of the development of conceptional plans for the conveyance of storm water through the mapping area to appropriate discharge points. Phase IV consisted of the preparation of final runoff computations and the preparation of a preliminary layout of feasible solutions to the problems: Cost estimates were also included in this phase. During the course of the study, meetings were held with Mr. Richard Straub, Weld County Engineer, to discuss the results of each phase. Through these meetings , various aspects of the scope of work were accented. These included the utilization of surface drainage to the maximum extent possible, the use of the 50-year storm as the major criteria for designing drainage facilities wherever possible, and placing major emphasis on the areas where flooding is most acute (Evanston and the western edge of Firestone) . -2- SECTION II STUDY AREA TOPOGRAPHY AND NATURAL DRAINAGE The towns of Firestone, Frederick, Dacono and the unincorporated Town of Evanston are part of a largely mining and agricultural area of Weld County, Colorado. The basin , consisting of 3,279 acres , or 5. 1 square miles (See Exhibit No. I ) of gently rolling land, is drained by a natural swale which has been improved by man to various degrees. This main drainageway basically flows northward between the Union Pacific Railroad and the towns of Frederick, Evanston and Firestone into a swampy area bordering Firestone Lake. From the lake area, the runoff eventually makes its way to the St. Vrain River. The drainage basin is traversed by two irrigation ditches . The Stanley Ditch crosses the basin in its extreme upper reaches. The runoff that could be diverted by the ditch is of insignificant size and would not effectively reduce the magnitude of the runoff in the basin' s lower reaches. On the other hand , the ditch banks are of such a size to indicate enough freeboard and little possibility of overtopping , thus eliminating the danger of transbasin diversion of storm runoff. The Sullivan Ditch has been rerouted underground with the abandoned part of the ditch filled in, thus eliminating any over- - topping danger or use as a runoff diversion canal . SUBBASINS The drainage basin has been divided into 26 subbasin for analysis and design purposes. Topographical features as well as the main drainage channel dictated the subbasin size and design points. All the basins are tributary to the main drainage channel along its course with the exception of the areas west of the Union Pacific Railroad. This area is effectively eliminated from the -3- remaining areas by the tracks until the main channel reaches its discharge point near Firestone Lake. Exhibit No. I shows the individual subbasins along with the general drainage routes. The subbasin designation was labeled such that A10, B10. . .etc. Represent design points along the main channel and All , B11 , etc. correspond to tributary subbasins to these main channel design points, respectively. -4- SECTION III HYDROLOGY STORM FREQUENCY Planning and design for urban storm runoff must be considered from the viewpoint of both the regularly expected storm occurrence, that is, the initial storm, and less regularly expected storm occurrence, that is , the major storm. The initial storm will have a frequency ranging from once in two years to once in ten years. The major storm will generally have a return period of 25, 50, or 100 years. The objectives of the major storm runoff planning and design is to eliminate major damage and loss of life. The initial drainage system is necessary to eliminate inconvenience, frequently recurring minor damage, and high street maintenance. For the purpose of this study, the magnitude of the peak rates of storm runoff and volumes of storm runoff at the various design points were determined for the 2, 5, 25 50 and 100-year storm frequencies. The 5-year and 50-year storm frequencies were then selected as the design frequencies for the minor and major storms. Major emphasis has been placed on the major design storm due to the lack of adequate major storm drainage facilities at present. HYDROGRAPH PROCEDURE For determining the magnitude of storm runoff flows there are several methods available for design purposes. The Colorado Urban Hydrograph Procedure (CUHP) , as presented in the "Urban Storm Drainage Criteria Manual", published by the Denver Regional Council of Governments (DRCOG) in March 1969, was selected. The CUHP procedure is based on the derivation of a synthetic unit hydrograph and statistically determined rainfall information. The unit hydrograph approach to a storm runoff determination was originally developed by Sherman -5- in 1932. The synthetic unit hydrograph, which is used for analysis when there is no rainfall -runoff data for the basin under study, was developed by Snyder in 1938. The unit hydrograph reflects the integrated effects of basin Para- ` meters such as area, shape, slope and ground cover, and defines the shape of the runoff hydrograph for one inch of direct runoff over the area. For storms where the runoff depth is different from the unit depth, the actual storm hydrograph is determined by multiplying the ordinates of the unit hydrograph by the actual runoff depth and adding to obtain the final storm runoff hydrograph. The basic premise of the unit hydrograph is that individual hydrographs result- ing from the successive increments of excess rainfall which occur throughout a storm period will be proportional in discharge throughout their length, and that when properly arranged with respect to time, the ordinates of the indi- vidual unit graphs can be added to give ordinates which will represent the total storm discharge. The hydrograph of the total storm discharge is obtained by summing the ordinates of the individual hydrographs. Basin parameters were determined from the available mapping. These parameters included the area, channel length, channel length closest to the basin centroid and the average channel slope through the basin. In addition, flow times along the channel through the basin were computed. Areas of actual and projected land use patterns were used in the determination of the various basin charac- teristics. Basin and hydrological characteristics included the percentage of pervious and impervious areas , infiltration rate in inches per hour, pervious and impervious retention depths , the impervious loss percentage, and hydrograph time and peak coefficients reflecting the type and degree of development. The -6- storm hydrograph coefficients , Cp and Ct, which reflect the basin slope and degree of perviousness or imperviousness , and degree of sewered area, are based on the values recommended in the "Urban Storm Drainage Criteria Manual ." In utilizing the CUHP method of analysis , the effective rainfall , or the amount of rainfall that can be considered as runoff , must be determined. The design rainfall was extrapolated from the "Urban Storm Drainage Criteria Manual . " This design rainfall is then modified by taking into account hydrological and basin characteristics such as perviousness-imperviousness , depression and detention losses , and infiltration losses to arrive at the effective rainfall . Pervious-Impervious areas All parts of a basin can be considered either pervious or impervious. The pervious part is that area where water can readily infiltrate into the ground. The impervious portion of a basin is the area that restricts infiltration such as buildings, streets , parking lots , and sidewalks. Thus, it can be seen that increased urbanization readily increases the amount of runoff. In order to accurately determine the amount of runoff, the amount of impervious areas must be calculated. In addition, a prediction of future urbanization must be made so as to provide a properly sized drainage system for a reasonable period of time. A review of aerial maps of the basin indicates a large amount of undeveloped land within the existing town limits. Thus, for the purpose of this report, runoff calculations were based on the future condition that all areas within the present town limits are completely developed (subbasin Alp, All , B10, B1l and part of Cll for Firestone; subbasins C10, 010, D21 and parts of C1l , DIl and D22 for Evanston; subbasins E10 and D23 for Frederick; and subbasins G10 -7- and H10 for Dacono) . All of the remaining areas were assumed to remain in an agricultural state. Values of impervious areas ranged from 2.5 percent for agricultural subbasins to 40 percent for urbanized subbasins. These values represent an average for the entire subbasin. Depression and Detention Losses Rainwater that is collected and held in small depressions and does not become part of the general surface runoff is called depression storage. Most of this water eventually infiltrates or evaporates. Detention losses include storm water intercepted by trees and bushes and water that is detained on the surface which does not run off until the storm is over. The values used for detention factors in this study were 0. 10 inches for impervious areas and 0. 30 to 0.40 inches for pervious areas in urban and farm areas, respectively. These values correspond to recommended factors presented in the "Urban Drainage Manual ." Infiltration Losses The flow of water through the soil surface is called infiltration. Soil type is the most important factor in determining the infiltration rate. Sandy soils normally have higher infiltration rates than do clay type soils . The ground cover also plays an important role in determining the infiltration rate. Other factors affecting infiltration rates include: slope of the land, temper- ature, quality of water, soil compaction, and saturation of the soil . Soil types in the Firestone-Frederick area were obtained from the U.S. Soil Conservation Service (SCS) maps. The soils are primarily of a sandy loam nature and are well drained. Infiltration rates of .5 inch to 1 .0 inches were utilized in the study as recommended by the SCS and as substantiated by field tests. -8- RUNOFF ANALYSIS The drainage basin was dividied into 26 subbasins and the 2, 5 25, 50, and 100-year storm hydrographs were computed for each subbasin. Lag times were computed to properly route the storm runoff through the drainage basin to arrive at accumulated storm hydrograph values at the various design points. These accumulated storm hydrographs represent values under existing conditions ; i .e. , with the existing drainage facilities. Although storm hydrographs were computed for the several storms listed above, the 5-year and 50-year storm were chosen as the design storms with major emphasis placed on the 50-year storm. — The following tables represent the various design criteria for each subbasin, the storm hydrographs for each subbasin and the accumulated storm hydrographs for the existing routing conditions. -9- o_ C E 0 \ i+ C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 • C •- \\\\\\ \ \ \ \\\\ \\\ \\\ \\\ \ \\\ ) v M Ma' M Ma' en M? en- a' en en Ma' en a' Ma' Ma' a'a" J 4) d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C 0 ♦J L fp C OOO OOOO OOOOOO O LA Ln LA LAO Ln Ln LA Ln LIO Ln L \ i� C 0 0 0 0 ^ 0 0 0 0 0 0 w v C OO MOOMO - MO ^ MO enOMO fneno en0 MMMM O. LA L!\l!\ Ln Ln Ln LA Ln LA Ln LA Ln Ln l!\l.fl l.fl Ln Ln LA In Ln LA LA in Ln t„) . 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 H W W OOMOO tnO ? MO •- MO t+1OMO MMO MO enenen +l x i-I MMa' MM,SMM? M? a' MSM? MS ? Mamaa- aa- N U N 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 > K fr W ¢ f W X CO N OLD OS rn r-- o-%co ^W CO CON ^ ON en•O LO LD N N N ^ NN N N 1.f1 N N ? O M a' N N M n N Ln N M I t\ n ¢ a E 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 H - -Cr W — J U in m — W ¢ (.'J P N.O CO ? N.Lc) S ? a' Lll Lfl- nNn Q1 LDS N O NLO N N N a- N J-7 N Ll% O M S Ol N n 1.Lf1? N.a- 1 O LA LC) M Lfl LO Ce N J`-' 0 0 0 0 0 0 0 0 ^ 0 0 0 0 0 0 0 0 0 - 0 ^ 0 0 0 ^ O Q ✓ m x CO O - z in N Q V m CO i O C •- O O Ln O O Ln O O Ln O Lll LA 0 0 0 LA O Ln LAO Lh O In Ln In Ln N U 7) O ON OO N O LA N O N N O LA O N O N N ONO N N N N L L s - a' a a' N a- ^ s S -1- o. D. E N ` Ot O LA O CA LA O ^ O Al Al OO O NO LIl OOOOO OOOO CO . . . . . . . . . . . . . . . . . . . . MO Ol N LC) O MN OICO cm f.? re) CI O1.S LAO O MO LOS In In L N LlL M^ M? ^ Lfl ^ ? O ^ n Al Ln 1? J LD Ol CO M U ^ ^ NLD ^ M^ M? ¢ _ C rp O N O ^ NO •- N O N N M J O O ^ O ^ O O ^ N N N N N ^ a Q ¢ Q m mmU 00000O00P wLL LL c cox x — ¢ w _ N 4- 0) 4 E r. 0 ^ L.. t V 1 01-7 N O- N.CO O N N S M O 01 1.— 0 N M O — N MOO O J O Q N N- - ? N ^ on N.N M 1/\^ N O N VO CO O CO M O ^VO VD N NfY > N —'-' — — J — N M cc 4- } s V- I N O Ln ^ U1 VO M S .-- M N 01 N O N U1 1� N? M U1 N Ltd O1 L11 VO 1� U1 M O 4) C 4- 111011 MIS? M -- Vpc0 M M L11 L11 — (7)-7 N MN N ? 0 C O 0 J U — — — N MMMM — N M ce 4- 4 E 0 J Li LINN N n 0l— L11 a-% co X Y O J O Q — U1— — M .- — MVO — N J — — CO N L11 01 1.11 N F- C0 >—' ^ M ^ — — .- N CL V- } S 4- RI 0 in ? MVO 01? NOl ^ MVO CO ? O OI-'VO - 01 CV n 1\Ol VO V0O � O W C 4- .4 CO M N VO Cr)N VO Ol M J VO M N M? O.T VO CO CO CO O N VO N L11 & fY U. - - NNNNN - - - M F- W 1L = 4- E N r 4 S 0 J L1 I�S - - N N VO VO - - - ? COO M S Vp M 0 1- 01 N ).0 Lf1 >- 0 J 0 < - ? - - M - - N .- - Non,- - n N 1.11 N M S VO LP? S O- Q F- K >v - N - - ^ N S N D N N 0- K 4 (1 } Y V r ce 4) 4- CO n N LAN N lJ'1 VO M M N N CO L/1 C N L Cl O O L11 M 1.11 CO O VO N •C O N 0- J U - N N N N N N - - 00 W 1' J U 0 - >- < C7 S F- O J t O D_' V- 0) CC O w E 0 J lL - 00 M n - M n Vp N O 0107 n J M O Vp U\N tom? n VO N Lf107 S N O D 0 Q - N 00 N 00 0 0 0 0 L11- MVO VO 01 CO O N N N - F- ce > - 2 N CC S N 4- O >- LO O a on?Vp L11? VO 111 CO VO Ol M-VO VD O al N O1 - N•0 NO.CO CO 1 C) CLF- eV - M - N CO - VO O1 f\ N nJ UlN M L1\ d J O - •- (l) LY - 4 (1) 4- Er 0 - L.4 CO O - LALn •- U\ - S 1--.4) N LN N 0-1 N VO A 01- -7 N in N CO N C 0 J 0 < O N 0 0 - 0 0 - 0 0 0 O O O M O N M- 1.? n - - O VD ♦- 1 > Ln CC Y } t 0 O V O O - 0 M - r•-• M 0 1 M N• N - N l U 1 1/4.0 O N NCO M (.0 N N 1 C 4-- - -- M �- N - - U1 - -7Ln- N VO N N N N N. N 0- J U — K - C f0 0 - N 0 -- NO .- NO - N - N M? O O - O - O - O - - Q Q Q 0 0 0 0 0 0 0 0 0 0 0 W U U O LO S 2 - Q LJ • N 4- O) 4- E.---, ^ CI Cl C' M 0 N Ol - M S LA a S n Z O S LL C C - I M O M 0 VD M ..O on N 0 N LA 1/40- - QJ - l0 n 00 I- J O Q M S N. M en ._ p I- K >v - - - .-- - - N — N K > 4- I Y w.". N O N O 0 CO N Ol CO COLA LA N 0 N- M N - N O - N CO O 2 N O W CH VD M Olen Vl l0 CO CO 01 N N. 00 N 0_ C U M S IN 0 N O a ^ - 4- 0) W E^ I- O 2 CO M N. - M S M O - N. M ,- S - Z O O LL . C C •O I N LA M CO S c- n en .-- •- M S 1p CTLA l0 H C --- •--- ,-- -O Q N m l .— M 0 0 2 - — — — 0 H >. ._ ._ — C CO _ K U- > Y4 H. - I moo S Ol in Ol Ito co — n ,r, lA LA ^ M N 0 M N- 0) d C H — 2 N N m CA IA 0 n M LA Vl r--. o 2 CO 0 LA 0_ 5 0 Men 2 N CO N 00 CA CA •— .— ,— — — LL N 4' E^ I\ N. M J M 0 en LA S CO N N N M M S N 2 O 0 LL . K C •O 1 N Ql S CT, S M ^- 1/40 M CI — N LA N. S 4\lIA H C O Q N N N CFI N en LA l - - - \ '0 H f C > ,-- - N C 4- > Y 4-.-- I f0 0 0 LA 01 S Al M N S N LA Ql M LA LA CO N - N LA MC 4. 1p O N. M - S N Ol 0 - M M LA IN LA n - N 0- 0 U Nen LA n n N. N. IN N - - - Ql H 0_• W W CO 2 CO O. W > Q - 4- 41 C C H 4- E^ N S S N N M CO CA •-• Ni 2 M n CO ? - M Z O J C C - I LA N M LA 00 N LA 0 l0 1/4.0 n .- Ni N M S Z O U H 0 >`Q-' - - M S' S LA LA 1/40 2 n N VI > Q N 2 LL Q 4- H Z W C Y 4 H. - - Q C U' > 10 00 S - N 0 N LA - Ol LA LA S - LA M S M LA M 0 0 0 I L C 4- LA IN - l0 S N O a - n n co - N M S CO W U H Z LA W v - - M M M M S S S S J - Q CO LOCI C Q 0 W 0 H J I- - O Q 0 C J Z 4- y C] O - Lo- E_ L^ - N. N 01 LA 0 00 O N MLA O N S LA O N 2 7 N C C I • • • • • • • • • • = O - L0 Cl l0 n 0 N S S LA O - - N Cl U - O O O a - - N N N M M S S -1- Z U X H C >" Q W CO N Q 4- CO C Y 4-.--. 0] > MI 00 N. S 0 S LA - en 0 LA LA N N I- LA S p - O I W C 4- ON Ol N Al S LA 10 1/40 LA LA LA - N en O CO N 0_ 0 U - - - N N N N N Al M C�. T L LA - en N. N N N N N - LA LA N CO N CO N O M en N N N 1/40 LA OO CO - LA l0 .- l0 Ol CO 0 0 y L S LA ON N. M CO CO CA •-• 2 n LAO IN N N. L U M S N. 1/40 n O N en S - - - - N Al Q Q — L - - N N N N M H • VI 0 0 0 7 O D D 9 C LL 0. a N O. 0. 10 0 ^ 0 0 0 0 N - N - N N C LL W N G -0 - Q U U CO L N .1-•.1-• O M C C C • C 0. .0 O - U S = 0 co 0_ 0. - 0_ U 0. CO X 0 CO 0 CO C C ? - - OI 01 • S T O OI O N S en ✓i L .- _- _- N N C N 0 - 0 .- 0 VI U VI rD 0 0 - d61 N O W W VI 0 D O p • 0 on VI CO M .0 O - 0 0 C C M C C .0 S M N N N N N .0 2 (j 2 2 N N M 0 O r0 0 N 0 0 - 00000 CO C O H rO M CO N N W 0 O O O OOE -0 Ed] NE -0 E W — O M N — CO 0 10 M N 1O O CO D — C N N N _ — co — E — W N in Vf 0 O N VI LA 41 < 0000 CU CD Q Q 10 E O ro C 0) ON. 0 - c O O., O C O 00 0 0 0 00 Al •_ — — m — 3 — — .— L-- .— N NN e- .— .— ^- N _ L O 2 U' CI_O O LL W0 0 CD Q 00000 m C Q 0 0. = 0 SECTION IV HYDRAULIC ANALYSIS - EXISTING FACILITIES MAIN CHANNEL As was stated previously, the main drainage feature of the study area is an earthen channel that begins along the south side of Frederick and then parallels the Union Pacific Railroad in a northerly direction along the towns of Frederick Evanston, and Firestone, respectively. The following discusses the capacities of the various existing facilities along the route of the channel (refer to Exhibit No. II ) . Beginning of Channel to Main Street Bridge, Frederick The main channel has its beginnings along the southern edge of Frederick where it intercepts all of the runoff from the south ( including Dacono) and the east. This runoff is forced west by the ditch in order to bypass Frederick. The ditch, for most of its length in this section , was constructed at a flat slope (0.002) and thus has a limited capacity (approximately 300 cfs) . This allows the ditch to pass a storm somewhere between the 2-year (224 cfs) and the 5-year (360 cfs) frequency. However, the Town of Frederick is protected by a berm and any ditch overtopping from a more severe- storm would inundate the farmland to the south causing minimal damage to Frederick. As the ditch approaches the Main Street bridge, it steepens in grade (0.009) and its capacity increases to approximately 870 cfs, near the expected 50-year runoff of 889 cfs. Main Street Bridge, Frederick The Main Street bridge over the main channel has a waterway opening of 11 .9 feet by 7 feet (width x depth) . The maximum capacity of this opening , without overtopping the road (Hw = 8.4 feet) , is 702 cfs. Thus , the structure is almost -10- capable of passing the 25-year storm (732 cfs) but falls far short of the 50-year storm (889) cfs) . At this more intense storm, the storm water can be expected to back up at the bridge, inundate the surrounding farmland, and then overtop the road causing minor damages in the vicinity. Main Channel Between Main Street and First Street, Frederick The channel in this section has a varying cross-section and a varying slope. The upper reaches , near the Main Street bridge, have a steep slope (0.009) and a capacity well in excess of 1 ,000 cfs; more than adequate to handle the 50-year storm (889 cfs) . However, in the lower half of this section , the slope flattens out to approximately 0.0055 with a corresponding capacity of 525 cfs. Bank overtopping in this area would cause damage to the farmlands and the western edge of Frederick. There also exists the ever-present danger of the berm — failing , resulting in major damages. First Street Bridge, Frederick The First Street bridge over the main channel has a waterway opening of 14 feet by 8 feet, sufficient to pass the 50-year storm without overtopping the road. However , to reach this capacity, the headwater on the upstream side of the structure must reach 8.4 feet to overcome the structure' s inlet losses. This is impossible under the present conditions since the channel ' s berm is not that high and also has been eliminated at the bridge in order to drain subbasin E10. Thus , even a 2-year storm will inundate the field to the east of the ditch while a 50-year storm might encroach into the Town of Frederick. Main Channel Between First Street Bridge, Frederick and the Railroad Bridge, Firestone The channel ' s cross-section, and consequently, its carrying capacity , deteriorate into almost nothing in this section. In effect, the channel -11 - flattens out into a meandering stream, thus allowing any significant runoff to inundate a wide area. This causes minimal problems in Evanston due to a lack of building in this area. However, this causes many flooding problems in the northwest corner of Firestone in the mobile home park. Union Pacific Railroad Bridge, Firestone This structure is capable of passing the 50-year storm with a headwater depth of approximately 5 feet. This depth could be reduced by cleaning out the silted up opening. Main Channel Between Railroad Bridge and Grant Avenue Bridge, Firestone This winding section of the channel suffers from a combination of a relatively — flat slope and a small cross-section, thus making it impossible to carry any reasonably large amount of storm water without flooding the surrounding areas . Grant Avenue Bridge, Firestone This structure has a reasonable width of 18.8 feet. However, its depth is only 4 feet and this , coupled with low available headwater, results in a capacity of approximately 375 cfs , less than the 5-year runoff at this point. A major restriction in this area is the existence of several buildings upstream of the bridge that would be subject to flooding at higher headwater allowances . Downstream of Grant Avenue Bridge Downstream of the Grant Avenue bridge the main channel empties into a large low lying area where the storm runoff spreads out before flowing northward to the St. Vrain River. In effect , this marshy area acts as a natural detention pond. -12- Summary As was shown above, the main channel has insufficient capacity to handle the 50-year storm and in many areas the 5-year storm. As a result , much flooding occurs in the surrounding areas with the degrees of damage depending on the type of development. In all cases , the flooding should be alleviated in order to reduce the potential damage and make the surrounding lands more beneficial to use. OTHER FACILITIES Dacono Area The Town of Dacono is presently served by a surface drainage system consisting of curb and gutters, cross pans and culverts. The outlet consists of several large culverts, 42 and 48 inches in size, that drain under Highway 52 into the farmland north of town. This system, adequate for minor storms, will also greatly reduce the potential damage from major storms. Major storms will basically follow the same general route with the excess flow overtopping the highway. The mobile home park, located in the eastern portions of the town , is served by an underground system with a 36-inch RCP as an outlet. This , plus street drainage, should effectively pass a major portion of its storm runoff without any problems. Area Between Dacono and Frederick A swale and small ditch in the field just north of Dacono collects all of the runoff from the lands to the south and discharges them into the main drainage channel just south of Frederick. This ditch is only adequate for minor flows with the major flows inundating the surrounding farmland. -13- Subbasin F10 The runoff from this agriculturally developed subbasin is collected in a ditch along the east side of Frederick and diverted to the main channel . No signifi- cant problems are anticipated from this runoff until it reaches the main channel . Subbasin E10 This low lying subbasin presently drains into the main channel on the upstream side of the First Street bridge in Frederick through a breach in the main channel ' s berm. By draining this area in this way, the runoff carried by the channel is allowed to backflow into the subbasin causing extensive flooding in the field and fringe areas of Frederick. Subbasins D23 and D,4 The runoff from these subbasins collects at the First Street and Oak Street intersection in Frederick, from where it flows northward into Evanston. The land in this part of Evanston has many low lying areas where water collects and other areas where the streets are higher than the surrounding land, thus blocking potential drainage paths. A 24-inch storm drain pipe was constructed from the First and Oak Street intersection west along First Street to divert some of this runoff to Subbasin E10. This is useful for minor storms but the pipe;s capacity is limited (19 cfs) and does not significantly help the situ- ation in Evanston during major storms. Subbasins D71 and D72 This part of Evanston suffers from a lack of drainage paths to pass the runoff from Subbasins D21 and D22 and the excess flow from Subbasins D�3 and D24. Thus , water is allowed to collect in this vicinity before it can drain away, infiltrate or evaporate. -14- Subbasins D11 and D12 The runoff from these subbasins collects in low lying areas between Johnson and Dunmire Streets. After reaching a certain depth, the water then overtops Dunmire Street and eventually makes its way to the main channel . Subbasins C11 and C12 The runoff from these subbasins is collected by a concrete lined ditch that transports the water from McClure Avenue and First Street in Firestone to the _ main channel . The ditch has sufficient capacity to pass the 50-year storm (102 cfs) from these subbasins. However , the ditch has not been maintained adequately and its capacity has been seriously reduced through siltation. The grading at the entrance is such that the runoff has some difficulty entering the ditch. Subbasin C10 The runoff from this subbasin flows directly into the main channel . There does exist a low lying area to the east of Dunmire Street that collects water until it overflows to the concrete lined ditch. This is a relatively minor problem due to the small tributary area to this low spot. Subbasins B11 and B12 The runoff from these subbasins collects in a low lying area , just west of First Street and Granville Avenue in Firestone, until the water level reaches a height that allows it to overflow to the main channel . The low lying area lies in an open field, thus minimal damage can be expected. Subbasin B10 This small subbasin lies along the main channel . Most of the runoff flows directly into the channel with some collecting in several low lying areas. This subbasin is largely undeveloped at present , thus minimal damage can be expected. -15- Subbasins All and Al2 The runoff from these subbasins collects in a low lying area in the vicinity of First Street and Wooster Avenue where the depth can build up to almost two feet before it can flow across First Street and make its way to the main channel . Subbasin A10 Most of this subbasin flows into the main channel , however, there do exist some low lying areas in the mobile home park where water can collect. A major problem for this area , and the mobile home park in particular, is the water carried by the main channel . This water overtops the channel ' s banks, thus — inundating the mobile home park. — Subbasins Ell and A71 Subbasins Ell (with the abandonment of the Sullivan Ditch) and A21 are effectively isolated from the remainder of the study area by the railroad tracks. The runoff does enter the main channel between the railroad bridge and the Grant Avenue bridge in Firestone. The area is undeveloped with many natural — depressions where water collects. Minimal flood damage can be expected to occur in this region. Summary As can be seen from above, the major problem areas exist primarily in the Town of Evanston, the E10 subbasin at Frederick, and in the vicinity of the mobile home park in Firestone. Proposed storm drainage facilities have been basically limited to these areas since the remaining areas are: (I ) adequately served by major drainage facilities ; (2) can expect minimal damage from flooding ; or (3) have minimal facilities that greatly reduce the storm drainage problems , particularly those expected from minor storms. -16- SECTION V PROPOSED FACILITIES - MAIN CHANNEL GENERAL As was shown in the previous section of the report, definite improvements are necessary in order for the main channel to carry the 50-year storm. Several ways of handling the runoff were investigated with the following two alter- natives judged the only feasible ones. Alternate No. 1 calls for the use of upstream and downstream detention ponds in order to reduce the runoff, reduce the required downstream channel sizes , and also to utilize the existing structures along the channel . Alternate No. 2 outlines the necessary facilities to pass the 50-year storm through the basin with no detention. Exhibit No. II summarizes the proposed improvements. ALTERNATE NO. 1 The first major improvement that is necessary under this alternative is the construction of a detention pond in Subbasin F10 (Station 90+00) . This pond would intercept, store, and discharge at a controlled maximum rate all of the 50-year runoff from the subbasins to the south and east of Frederick. By restricting the discharge from the pond to a certain rate, which is much less than what the uncontrolled runoff would be, all downstream facilities can be reduced in size, thus lowering capital costs and requiring smaller rights-of- way. In addition, velocities in the downstream channel are usually lower, thereby reducing the potential for channel scour. Furthermore, a greater safety factor is present with the reduced flows and the potential for flow regulation at the pond. Future upstream urbanization can be more readily accommodated by the simple task of enlarging the pond size. The basic speci- fications for the pond are: -1 7- Controlled Discharge Rate 200 cfs Storage Volume 57 5 acre-feet Surface Area Required 11 5 acres (approx. 900' x 550' ) Outlet Invert Elevation 81 0 Top of Berm 89 0 (including allowance for 2' freeboard) Average Water Depth 6 feet In addition to the above, the pond should be provided with an emergency spillway in order to protect the pond from failure due to overtopping during an outlet blockage or a more intense storm occurring (100-year or greater) . The land around the pond should be graded in such a way to insure all runoff will enter the pond at the appropriate places. The exact pond shape and size should be chosen at the final design stages to minimize earthwork cut and filling . The pond will drain completely after each storm. The existing channel between the pond and the Main Street bridge in Frederick (Stations 90+00 to 79+40) should be cleaned out and regraded as necessary to insure a uniform grade from the pond. The existing ditch has enough capacity to handle the 200 cfs pond discharge, therefore, the necessary improvements in this section should be minimal . With the reduction in flow, the Main Street bridge (Station 79+40) has enough capacity. The downstream side of the bridge should be protected by either riprap, gabions , or a concrete apron , against erosion under high flows. This is especially true where the channel bends and heads north just downstream of the bridge. The existing channel between the Main Street bridge and the First Street bridge in Frederick (Stations 79+40 to 55+80) again has sufficient capacity to pass the reduced flow (200 cfs) . The channel should be cleaned and straightened in this area as necessary. -18- At the First Street bridge in Frederick (Station 55+80) , the only improvements necessary include rebuilding of the channel 's upstream embankment to keep the flow in the channel and erosion protection on the downstream side of the bridge. The upstream berm should be constructed to an elevation of 69.0 to allow enough headwater to pass the flow through the bridge. This entails additional facil - ities to properly drain Subbasin E10. These facilities are outlined in Section VI of the report. _ The existing channel between Stations 55+80 and 25+00 should be improved by constructing a grass lined channel with a bottom width of 10 feet , a depth of 4 feet , a side slope of 4: 1 , and a top width of 42 feet to handle the expected runoff. It is in this stretch where the runoff into the channel increases greatly from the influence of tributary areas. The design flows at points E10' D10, and C10 are 228 cfs, 447 cfs , and 577 cfs , respectively. By holding the ditch depth to 4 feet in order to drain the surrounding land, balancing cut and fill as much as possible, and varying the slope of the channel , the channel capacities are 400 cfs between Stations 55+80 and 50+00 (slope = 0.002) , 830 cfs between Stations 50+00 and 40+00 (slope = 0.008) , and 600 cfs between Stations 40+00 and 25+00 (slope = 0.004) . The average permanent right-of-way width required in this section, allowing for an access road along the ditch, is 60 feet. Since the channel 's flow has increased significantly by Station 25+00, a detention pond should be constructed between Stations 25 and 13. The benefits from this pond are the same as that mentioned previously for the pond at F10 with the added protection to the mobile home park in Firestone as an insurance factor. The pond specifications include: -19- Controlled Discharge Rate 200 cfs Storage Volume 33 5 acre-feet Surface Area Required 8 acres (approx. 1200 ' x 250' ) Outlet Invert Elevation 44 5 _ Inlet Invert Elevation 48 0 Top of Berm 53 0 The reason for the long, narrow shape of the pond is to minimize the amount of valuable land that is needed for the pond and the drainage channel . A grass lined channel should be constructed between the detention pond and the Union Pacific Railroad trestle (Stations 13+00 to 8+40) . A channel with a bottom width of 10 feet, depth of 3 feet , a side slope of 4: 1 results in a capacity of 290 cfs at a slope of 0.003, adequate to handle the pond discharge and the runoff from Subbasins A10, All , and A12 (290 cfs) . A right-of-way equal to 50 feet is anticipated. The invert of the railroad trestle should be lowered from elevation 45 to 43. This will improve the capacity of the structure and should be possible since a large amount of silt has accumulated under the trestle over the years. The invert of the channel under the trestle should be paved to improve the hydraulics . Due to the number of curves in the existing channel between the railroad trestle and the Grant Avenue bridge (Stations 8+40 to 0+00) and the shallow depth avail - able, a concrete lined channel is recommended for this section. This controls erosion, insures a shallower depth, and can easily be transitioned to the Grant Avenue structure. The ditch specifications are: Bottom Width 8 feet Depth 4 feet Side Slopes 1 1/2 : 1 Top Width 24 feet (includes a 2 foot horizontal lip on both sides to control siltation and lining undermining) -20- Required Right-of-Way Width 40 feet Slope 0.0025 Capacity 520 cfs Lining Thickness 4 inches The proposed ditch has been sized to handle the additional runoff from Subbasins A21 and Ell . The total runoff at this point is 528 cfs. At the location where the open channel crosses the proposed county road (Station 7+90) , it is recommended that the channel cross-section be kept and the road bridged over it. A simple and effective way to do this is by constructing a bridge of prestressed twin concrete "T" beams on cast-in-place concrete abut- ments. A span of approximately 24 feet is required. The concrete lined channel should be transitioned to the Grant Avenue bridge to allow the water to flow through the bridge without an appreciable headwater build up. The downstream side of the bridge should be protected from erosion in the transition from a concrete lined channel to the existing grass lined channel that flows towards Firestone Lake. ALTERNATE NO. 2 The first improvement for this alternate is to widen the channel along the south side of Frederick to handle the 50-year storm. A bottom width of 10 feet, depth of 5.5 feet and side slopes of 4: 1 results in a channel capacity of 850 cfs at a slope of 0.002. The Main Street bridge (Station 79+40) has an existing capacity of approxi- mately the 25-year storm. Therefore, it is recommended that the structure be retained and allow the road overtopped during the 50-year or greater storm. Minimal damages can be expected to occur under these conditions. The down- stream side of the bridge should be protected from erosion. -21- The existing channel between the Main Street and First Street bridges in Frederick should be improved to handle the 50-year storm. This requires a channel with a bottom width of 6 feet, a depth of 6 feet and side slopes of 2: 1 . This gives a channel capacity of 990 cfs at a slope of 0.007. The reason for the steep side slopes is the narrow right-of-way available. The required right-of-way is 40 feet. — The upstream channel berm at the First Street bridge (Station 55+80) should be built up to elevation 71 . 5 feet to allow the structure to pass the 50-year storm without inundating Subbasin E10. Section VI outlines the required improvements necessary to drain Subbasin E10. The downstream side of the structure should be protected from erosion. The existing channel between the First Street bridge in Frederick and the rail - road trestle (Stations 55+80 to 8+40) should be improved to a section with a bottom width at 15 feet, a 5 foot depth, and side slopes of 4 : 1 . This section , at a minimum slope of 0.003 , results in a capacity of 1 ,000 cfs , more than adequate to pass the 50-year runoff of 889 to 967 cfs. The permanent right-of- way width needed is 75 feet. The channel should be located so as to minimize cut and fill requirements and also to allow the surrounding land to drain into the channel . The invert of the railroad trestle at Station 8+40 should be lowered by 2 feet to elevation 43.0 to increase its capacity. The invert should be paved as outlined in Alternate No. 1 . The channel section between the railroad trestle and the Grant Avenue bridge (Station 8+40 to 0+00) should be concrete lined for the same reasons as out- _ lined in Alternate No. I . The channel specifications include: -22- Bottom Width 20 feet Depth 4 feet Side Slope 2: 1 Top Width 40 feet (includes 2' horizontal lip) — Required Right-of-Way Width 65 feet Slope 0.0025 Capacity 1 ,140 cfs Lining Thickness 4 inches The proposed ditch has been sized to handle the additional runoff from Subbasins A21 and Ell (total of 1 , 107 cfs) . At the location where the open channel crosses the proposed county road (Station 7+90) , it is recommended that the channel cross-section be kept intact and the road bridged over it. Again, a prestressed twin concrete "T" beam bridge, spanning approximately 40 feet , is recommended. Due to the inadequacy of the Grant Avenue bridge to pass the 50-year storm and the lack of available headwater, it is recommended that the Grant Avenue bridge be replaced with a new concrete "T" beam bridge which would span the concrete lined channel . This will allow the structure to cause no impedance to the flow, thus keeping the water depth within the maximum allowed by the surrounding ground conditions, four feet. The downstream transition from the concrete lined channel to the existing grass channel should be designed to minimize the potential for erosion. ADVANTAGES AND DISADVANTAGES Both of the above discussed alternatives are pliable solutions to the main channel drainage problems. However, there are many advantages and disad- vantages associated with both solutions that should be pointed out. The advantages of the detention pond system, Alternate No. 1 , over the open channel system without detention, include: -23- I . The reduction of the high, instantaneous peaks of runoff into smaller, longer peaks. This , in effect, utilizes the maximum size of the structure for longer periods of time, thus more easily justifying the capital costs of the facilities. 2. The reduction of the runoff, by the semi -regulation of the pond , results in: a. Smaller downstream facilities. b. Usually lower water depths with smaller flow velocities , hence less possibilities for scour. c. The retention of all of the existing bridges since they are more capable of passing smaller flows. d. Increased safety factor downstream through smaller flows and velocities, especially in the Firestone mobile home park area. e. Lesser right-of-way requirements for downstream facilities. f. Lesser maintenance requirements on the smaller facilities. g. The reduction of any shock loading on the channel 's discharge point, the marsh area around Firestone Lake. The area can more readily accept and pass along downstream the same quantity of water in smaller quantities spread out over a longer period of time. 3. The ease in expanding the pond size to accommodate additional runoff from unanticipated urbanization. 4. The utilization of two ponds has a greater effect on flow reduction for the entire channel length rather than the use of one pond which would require approximately the same area but would also need larger channel sizes to handle the flows. -24- 5. The potential for combining the use of the ponds as a recreational area and a detention pond. The disadvantages of the detention pond system over the open channel system without detention are: 1 . The large areas required for the ponds. 2. The higher capital costs due primarily to the cost of pond construction and land costs. (This is offset somewhat by the savings in smaller down- stream facilities , narrower downstream rights-of-way, and the retention of all existing structures. ) 3. The maintenance requirements of the pond. This is offset somewhat by the reduction in maintenance on the main channel due to smaller facilities. In any case, detention pond or grassed open channel , the maintenance can be reduced appreciably by the use of low maintenance dryland types of ground cover such as buffalo grass or crested wheat. OTHER ALTERNATIVES Other alternatives for solving the main channel problems were investigated. In general , these alternatives were eliminated because they were more costly, involved duplication of required facilities, or were not as practical as Alternates 1 and 2. These other solutions included: 1 . The use of extremely wide channels , utilizing the natural topography as much as possible to carry the flows , thereby lowering channel construction costs. This does not solve headwater problems at the existing structures , and requires a large amount of right-of-way acquisition. -25- 2. The diversion of the runoff at Frederick to the west side of the railroad tracks. This requires a large ditch on the west side of the tracks and another ditch on the east side to intercept the runoff from Evanston and Firestone, a more costly solution. 3. The utilization of detention ponds at other locations, or the use of only one pond. These alternatives did not appear to be as practical in saving the existing bridges nor have as much of a safety factor in protecting the mobile home park in Firestone. -26- SECTION VI PROPOSED FACILITIES - TRIBUTARY AREAS GENERAL This section of the report presents solutions to the drainage problems that exist in the main channel ' s tributary areas as outlined in Section IV. Major emphasis again was given to providing facilities to handle the 50-year storm with minimal damage. Exhibit No. II outlines the proposed facilities as discussed below. SUBBASIN E10 The need for this area is to drain the subbasin into the main channel without having the channel ' s water backflow into and inundate the subbasin. The lack of an appreciable elevation difference between the channel and the farmland in the lower reaches of the subbasin necessitated the construction of a high channel berm to keep the channel 's water inside the channel . Attempting to drain the subbasin into the channel on the upstream side of the First Street bridge, by means of culverts, would cause backwater problems due to the high water surface in the channel . It is recommended that culverts be installed _ under First Street to outlet into the channel on the downstream side of the bridge where the water depth is appreciably lower. This would limit the back- _ water effect to more tolerant levels. At the final design stage, more accurate water surface elevations should be determined to substantiate this fact. The 50-year runoff at this point is 120 cfs , 101 cfs from Subbasin E10 plus 19 cfs diverted from Subbasin 023 through the 24-inch storm drain pipe along First Street. Two 48- inch CMP culverts (or one 4' x 6' concrete box culvert) will pass this flow at a headwater to depth ratio of 1 .0. This will result in some water backup into the field (to elevation 67.0) but should cause minimal damage. -27- As outlined in the next subsection, it is desirable to divert all of the runoff from Subbasins D�3 and D24 to Subbasin E10. This will result in a additional 120 cfs flowing down First Street, approximately half on each side of the street. Thus , 60 cfs can be diverted directly into the main channel on the north side of the bridge while another 60 cfs will collect on the up- stream side of the culverts (south side of the street) . This additional run- off can be handled by adding another 48-inch culvert to the two already proposed , (or enlarging the concrete box culvert to 4' x 9' ) thus keeping the headwater to depth ratio at 1 .0. The main channel improvements are sufficient to handle this additional flow. SUBBASINS D23 AND D24 The runoff from these subbasins collects at the intersection of First Street and Oak Street in Frederick, from where it flows north to cause immense problems in Evanston. The existing 24-inch diversion pipe to Subbasin E10 along First Street is only adequate for carrying nuisance water and runoff from minor storms. To alleviate the Evanston problem, it is desirable to divert all of the runoff west into Subbasin E10, a more direct path to the open channel . One method of solving the problem is to install an underground pipe system to divert the water. This would require an extensive, and expensive, pipe network throughout Frederick in order to collect the runoff. At best , this system would only intercept part of the runoff , since pipe systems are normally designed for minor storms , with the remainder still flowing into Evanston and causing problems. A more logical solution is to provide a surface path, to carry the runoff from major storms , from the intersection west along First Street. This would involve -28- the removal and lowering of First Street between Oak and just west of Elm, the installation of curb and gutter between Oak and Elm, and the construction of roadside ditches (60 cfs capacity) on both sides of the street between Elm and the main channel . The roadside ditch is sized as follows : Bottom Width 4 feet Depth 2.5 feet Side Slopes 2: 1 Top Width 14 feet Slope 0.002 Capacity 60 cfs By providing approximately a half foot difference between the street crown and the ground at the right-of-way line, will provide enough capacity to pass the 50-year storm (134 cfs) with minimal damage. The curb and gutter sections would not quite carry the 2-year runoff, thus future consideration could be given to the installation of a pipe system to further refine the system. This diversion system removes the bulk of the runoff from entering Subbasin D22 (Evanston) , thereby greatly decreasing the size of the necessary drainage facilities there. SUBBASINS D21 , D22, D11 , AND D12 The following table shows the decrease in the runoff in these basins by diverting the 023 runoff to the west: SUBBASIN D21 SUBBASIN D22 Storm Frequency w/Diversion w/o Diversion w/Diversion w/o Diversion 2-yr. 11 cfs 39 cfs 2 cfs 33 cfs 5-yr. 16 cfs 62 cfs 6 cfs 55 cfs 25-yr. 32 cfs 116 cfs 22 cfs 114 cfs 50-yr. 41 cfs 138 cfs 29 cfs 136 cfs 100-yr. 51 cfs 162 cfs 37 cfs 161 cfs Since the accumulated flow at the discharge point of Subbasin D21 has been substantially reduced, the runoff can be diverted north along Dunmire Street -29- to Subbasin D11 . This eliminates the need for an additional drainage path to the main channel from Subbasin D21 . The accumulated runoff at design point D11 from Subbasins Dii , D12, D21 , and D�3 is 14, 27, 68, 89 , and 112 cfs for the 2 , 5 , 25, 50, and 100-year storms , respectively. The amount of development along Johnson and Dunmire Streets makes the use of roadside ditches impractical . Instead, it is recommended that the runoff be handled by a surface system consisting of a concrete gutter pan on both sides of Johnson and Dunmire Streets. This would involve the construction of a 6-foot asphalt shoulder between the existing paved streets and the proposed pan (see detail , Sheet No. 2, Exhibit No. II ) . This will save the pavement edge from breaking up and allow the pan to be constructed lower, thus increasing the street capacity to transport the runoff and free the traffic lanes from carrying water during minor storms . Concrete cross pans should be installed at various points to allow the runoff to make its way to the main channel . Roadside ditches are recommended along the gravel cross streets (Williams , Koch, and Hedges) primarily due to the lack of driveway entrances along the streets. Culverts have been proposed at Hedges and Dunmire because the flat grade of Hedges Avenue makes it impossible to grade the ditch in order to transition to a cross pan. The culverts (24-inch CMP) are sized for the 5-year storm with the major storm excess flowing over the street. The grass lined ditch along Johnson Street just north of Hedges Avenue has been proposed to drain the low point in that area, which is approximately one foot lower than the street inter- section. This would require culverts ( 18") to be installed at all driveways. This street surface drainage system discharges into a grass lined channel at Dunmire Street and Hedges Avenue. The ditch , which drains into the main channel , -30- has a bottom width of 4 feet , a depth of 2.5 feet , side slopes of 4: 1 , and a capacity of 120 cfs at a slope of 0.003. The required right-of-way for this ditch is 40 feet. An alternate solution to these subbasins would be the construction of curb and gutter for all the streets. This would involve the paving of the gravel streets and additional parking lanes along the paved streets, thus making it more expensive than the above-mentioned system. SUBBASINS C10, C11 , AND C12 The existing concrete lined channel that drains these subbasins should be cleaned out and repaired as necessary. In addition, the grading at the inlet should be improved so that it can accept the runoff more readily. A low point exists on Dunmire Street just south of McClure Avenue. This area can be provided with concrete gutter pans and cross pans as outlined in the previous subsection. SUBBASINS B10, B11 , AND B12 The runoff from these subbasins that collects at First Street and Granville Avenue in Firestone (60 cfs) should be transported to the main channel (or detention pond if that alternate is selected) by means of a grassed channel . The capacity of this channel at a slope of 0.008 is 75 cfs with a bottom width of 2 feet, a depth of 2 feet, and side slopes of 3: 1 . The required right-of-way for this ditch is 30 feet. SUBBASINS A10, All , AND Al2 There basically exists two major problems for these subbasins ; the depression at the lower end of Subbasin All near First Street and Wooster Avenue, and a means of outletting the water into the main channel without allowing the back- water from the channel to inundate the mobile home park (Subbasin A10) . -31- The problem with the depression at Wooster Avenue and First Street is that it lies approximately 2 feet lower than the surrounding land. The land to the north is much higher and an appropriate discharge point lies relatively far away, thus making a surface drainage system in this direction expensive and somewhat impractical . The more obvious drainage path is west through the mobile home park. However, the amount of development has restricted the available right-of-way width, thus making an open grassed channel impractical and dangerous. In addition, the relationship of the depression's elevation (45.8) to that of the main channel (proposed 43.0) does not allow enough cover or slope to install a pipe system that could transport even the 2-year storm. A grate covered concrete lined channel is proposed to drain this area (see detail , Sheet No. 1 , Exhibit No. II ) . The route is along the shoulder of a gravel road through the mobile home park. By making the walls vertical and covering the channel with a grate, the facility can be constructed in a narrow area, can support traffic loadings, and provide a safe structure without impeding the needed capacity. In addition, the channel can provide a mean? of draining some of the low lying areas in the mobile home park. The second problem is how to allow the runoff from these subbasins to enter the main channel and minimize the backwater effect. The anticipated water depth on the west side of the railroad tracks is 4 feet with the east side depth somewhat higher. It is recommended, at this preliminary stage, that the water be discharged into the channel on the west side because the backwater effect will be less. A 42-inch culvert should be bored under the railroad to drain the area (at final design analysis, it might be feasible to divide the railroad trestle opening into two separate sections to carry this runoff and the main channel runoff, or it might be feasible to discharge the runoff from -32- these subbasins into the channel on the east side of the tracks) . The natural depression between the tracks and the mobile home park should be used as a small backflow and retention pond (capacity, 1 .2 acre-feet ; surface area , 1 acre; maximum depth, 4 feet) for the runoff from the concrete ditch from Subbasin All and from the surrounding area. This would prevent water from backflowing into the mobile home park; however, some water may collect at Wooster and First during major storms. -33- SECTION VII COST ESTIMATES GENERAL Estimated costs for construction, engineering , legal , and miscellaneous costs are included in this section. Construction costs are based on the use of materials and techniques indicated in the report. Alternate types of materials for various items often exist, and some savings may result by careful evaluation at the final design stage. Due to the preliminary and conceptual nature of the study, the estimated costs are based on approximate quantities and should be analyzed with this fact in mind. UNIT COSTS The unit costs for major items that have been used for estimating purposes are indicated in the following table. These costs were arrived at through con- sultation with contractors , suppliers , and recent construction bids for similar projects. Additional sources of information included the latest copy of the Colorado Department of Highways Cost Data book, the Dodge Estimating Guide, and Mean' s Building Construction Cost Data. The costs are approximate in nature but should be representative of late 1975 construction costs. Excavation , including minimal haul , water and compaction $ .75/C.Y. Seeding and Mulching S00.00/acre Channel lining , concrete, installed 100.00/C.Y. Structural concrete, installed , including reinforcing 125.00/C.Y. Concrete curb and gutter 4.50/L. F. Concrete gutter pan 3.25/L. F. 18" CMP culvert , in place 12. 50/L. F. 24" CMP culvert, in place 17.50/L. F. 42" CMP culvert, in place 25.00/L. F. 48" CMP culvert, in place 30.00/L. F. Pavement removal . 50/S .Y. Concrete removal .75/S. F. Gabions 50.00/C.Y. 2" asphalt cement pavement 2.00/S.Y. -34- 4" aggregate subbase $ 1 .00/S.Y. Subgrade preparation •75/S.Y. Prestressed concrete "Twin T" beam bridge, installed, including abutments, wearing course, railings 12.50/S. F. -35- COST SUMMARIES MAIN CHANNEL - ALTERNATE NO. 1 1 . Transition from concrete lined channel to grassed channel , Station 0+00 $ 1 ,500.00 2. Concrete lined channel , 0+00 to 8+40 28,000.00 3. New bridge at proposed county road , 7+90 8,250.00 4. Grass lined channel , 8+40 to 13+00 750.00 5. Detention pond , 13+00 to 25+00 27,000.00 6. Grass lined channel , 25+00 to 55+80 10,000.00 7. First Street bridge improvements , 55+80 1 ,500.00 8. Improvements to existing ditch, 55+80 to 90+00 1 ,500.00 9. Detention pond, 90+00 42,000.00 Total Construction Cost $120,500.00 15% Contingencies , including Engineering , Administrative, and Legal Costs 18,075.00 TOTAL PROJECT COST $138,575.00 SAY $139,000.00 PRELIMINARY LAND ACQUISITION AND RIGHT-OF-WAY NEEDS Station 0+00 to 8+40, 40' width 0 8 Acres Station 8+40 to 13+00, 50' width 0 5 Acres Pond 13+00 to 25+00 8 0 Acres Station 75+00 to 55+80, 60' width 4 3 Acres Station 55+80 to 90+00, 0' width 0 Acres' Pond 90+00 11 5 Acres TOTALS 25. 1 Acres *Existing ditch capacity adequate, no additional right-of-way needed. -36- MAIN CHANNEL - ALTERNATE NO. 2 1 . Transition from concrete lined channel to grassed channel , Station 0+00 $ 2,000.00 2. Removal and replacement of Grant Avenue bridge Station 0+00 15,000.00 3. Concrete lined channel , 0+00 to 8+40 43,000.00 4. New bridge at proposed county road 13,000.00 5. Grass lined channel , 8+40 to 55+80 22,000.00 6. First Street bridge improvements , 55+80 2,000.00 7. Grass lined channel , 55+80 to 79+40 7,000.00 8. Grass lined channel , 79+40 to end 5,000.00 Total Construction Cost $109,000.00 15% Contingencies, including Engineering , Administrative, and Legal Costs 16,350.00 TOTAL PROJECT COST $125,350.00 SAY $126,000.00 PRELIMINARY RIGHT-OF-WAY NEEDS Station 0+00 to 8+40, 65' width 1 3 Acres Station 8+40 to 55+80, 75' width 8.2 Acres Station 55+80 to 79+40, 45' width 2 4 Acres Station 79+40 to 95+00, 70' width 2 5 Acres TOTAL 14.4 Acres -37- TRIBUTARY AREAS Subbasin E10 Culverts at First Street Bridge $ 4 ,000.00 Subbasins 023 and D24: Pavement removal $ 850.00 -- Concrete removal 900.00 Concrete gutter pans 260.00 Curb and gutter 3 ,825.00 2" Asphalt cement pavement 4,200.00 4" Aggregate base course 2,100.00 Subgrade preparation 1 ,500.00 Roadside ditches 250.00 Subtotal $ 13,885.00 Subbasins D11 , D12, D21 and D22: Pavement removal $ 500.00 2" Asphalt cement pavement 8,200.00 4" Aggregate base course 4,100.00 Subgrade preparation 3,050.00 Concrete gutter pans 20,000.00 Roadside swales 1 ,000.00 24" Culverts 1 ,400.00 Grass lined channel 650.00 Subtotal $ 38,900.00 Subbasins Cip, C11 , and C12: Clean and repair existing concrete channel $ 2,500.00 2" Asphalt cement pavement 1 ,100.00 4" Aggregate base course 500.00 Subgrade preparation 400.00 Concrete gutter pans 3,000.00 Subtotal $ 7,500.00 Subbasins 610, 811 , and B12: Grass lined channel $ 500.00 Subbasins A10, A11 , and Al2 : Concrete channel $25,000.00 Detention pond 2,500.00 Subtotal $ 27,500.00 Total Construction Cost . . $ 92,285.00 15s Contingencies , including Engineering , Administrative, and Legal Costs . . . $ 13,842. 75 TOTAL PROJECT COST . . . S106, 127.75 SAY . . . $107,000.00 -38- PRELIMINARY LAND ACQUISITION AND RIGHT-OF-WAY NEEDS Subbasin D10 , 35' width, 525' length 0. 4 Acres Subbasin 610, 25' width, 725' length 0.4 Acres Subbasin A10, Pond 1 .0 Acres TOTAL 1 .8 Acres -39- SECTION VIII SUMMARY, CONCLUSIONS , AND RECOMMENDATIONS SUMMARY AND CONCLUSIONS 1 . There exist major drainage problems in the Firestone-Evanston-Frederick area that need immediate attention. 2. Future urbanization will increase the magnitude of these problems. 3. The Stanley and Sullivan Ditches were investigated and then eliminated as possible facilities for diverting the runoff. They were also eliminated as a possible source of runoff from other basins. 4. Extensive hydrologic studies were undertaken to define the subbasin's design parameters. 5. Preliminary runoff computations were made at selected design points to define the drainage problems. The runoff calculations were made for the 2 , 5 , 25, 50, and 100-year storm frequencies. 6. The major problem areas include the main channel between Firestone and Frederick, the mobile home park in northwestern Firestone, the southern half of Evanston , and the northwest corner of Frederick. 7. The 50-year storm was selected as the design storm for proposed major facilities. 8. Two possible solutions were presented to solve the existing main channel 's drainage problems. Alternate No. 1 consisted of two detention ponds connected by open channels. Alternate No. 2 consisted of an open channel system sized to pass the 50-year storm without detention facilities. -40- 9. Solutions were presented to solve the drainage problems in the main channel ' s tributary areas. Surface drainage facilities were selected wherever possible. 10. The main channel ' s Alternate No. 1 , although more costly and requiring more land, appears to be more feasible from an engineering and safety point of view. The use of detention ponds allows the use of smaller down- stream facilities and the use of existing facilities. The regulation of pond discharges can provide a higher factor of safety for downstream facilities. 11 . The estimated project costs including contingency items are : Main Channel , Alternate No. 1 . . . . $139,000.00 Tributary Areas 107,000.00 Total $246,000.00 12. The land acquisition and right-of-way needs are: Main Channel , Alternate No. 1 . . . . 25. 1 Acres Tributary Areas 1 .8 Acres Total 26.9 Acres 13. The facilities were designed for a future condition of all areas inside present town limits being fully developed and all areas outside these limits remaining in an agricultural state. It is recommended that if any of these agricultural areas experiences urbanization, the runoff from these areas should be restricted to the "historic runoff." This will insure that all of the proposed improvements presented in this study will not be overloaded and cause additional flooding problems. The regu- lation of runoff from these new areas can be restricted to the "historic runoff" by means of on site detention facilities. -41 - 14. The towns of Firestone, Evanston , Frederick and Dacono all contribute to the amount of storm runoff and all should help in the financing of the proposed improvements. RECOMMENDATIONS In recognition of the drainage problems in the Firestone-Frederick area and the magnitude of the necessary solution thereto, we offer the following recom- mendations: 1 . That the facilities outlined in the Alternate No. 1 - Main Channel and the Tributary Areas sections of the report be selected as the way of solving the existing drainage problem. 2. That the preparation of detailed plans and specifications be undertaken immediately. 3. That preliminary land acquisition negotiations begin immediately. 4. That the towns of Firestone, Frederick and Dacono, along with Weld County proceed to plan, finance and construct, with sound fiscal planning and sound public policy, the proposed improvements as set forth in this report . -42- Hello