Loading...
HomeMy WebLinkAbout1986-05-20; City Council; Resolution 8569* I i h 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 0 RESOLUTION NO. 8569 0 A RESOLUTION OF THE CITY COUNCIL OF THE CITY OF CARLSBAD, CALIFORNIA, ADOPTING THE BUENA VISTA LAGOON WATERSHED SEDIMENT CONTROL PLAN. The City Council of the City of Carlsbad, resolves as follows: 1. The Buena Vista Lagoon Watershed Sediment Control Plan prepared by lune Applegate and Associates September 3, 1985 indicated as Exhibit 3, ittached hereto and incorporated herein, is adopted as the instrument to guide .he control of sedimentation entering the Buena Vista Lagoon. 2. The Mayor is authorized to request funding from the State Coastal :onservancy to provide funding for construct ion of the recommendat ions :ontained in the report. 3. The Mayor is authorized to urge the members of the Buena Vista Joint 'owers Committee to approve an equitable arrangement of costs for an operation md maintenance program prior to implementing recommendations of the report. PASSED, APPROVED AND ADOPTED at a regular meeting of the Carlsbad City :ouncil held on the 20th day of May , 1986 by the following vote, to tit: - AYES: Council Members Casler, Lewis, Kulchin and Pettine NOES: None ABSENT: Council Member Chick c MARY H. JCASLER, Mayor \TTEST: (SEAL ) BUENA VISTA LAGQBN WATERSHED SEDIMENT CONTROL PLAN Prepared by June Applegate & Associates Philip Williams & Associates BUENA VISTA LAGOON AND WATERSHED SEDIMENT CONTROL STUDY FOR THE CALIFORNIA COASTAL CONSERVANCY Coastal Conservancy 205 (J) grant Buena Vista Lagoon Sediment Management Phase I I 83-058-81 -48-C BY JUNE APPLEGATE, P.E. September 3, 1985 2531 STATE STREET, CARLSBAD, CALIFORNIA 92008 (619) 729-7109 r Peter Grenell, Executim Officer Alp Jacobson, Bihancemnt Program mger Laurie Marcus, hroject Manager JUNE APpLEzAm & ASSOCIATES June -legate, Principal Philip williams, Principal Jane KerLinger, Hydrologist I, 11 . 111. Iv. V. VI. VII. "M CNE suhbasin characteristic Calculations APPENDIX m Cost Effectiveness Calculations 24PPENDIX lnREE APPENDIX FOUR fagoon mtation Calculations upper Middle Reach Erosion Calculations FIGURF: 2 subareas FIGUKE 3 Nawal conditions FIGURE 4 Detention Basin Iocations FIGUKE 5 FIGURE 6 Hydrographs FIGURE 7 TAELFll Erosion control Alternatives TABLe2 Sediment Size Distribution for TABLE 3 sediment source slmmccy hrdposed Channel Cross Section -tic Diagram of Bum Vista Lagoon Different Hydrographs tinva to peak for future conditions at the lagoon Used in Analysis TABLE 4 surmrary of NEC-1 peak flows and TABLE 5 Projected Sediment Reductions TABLE6 sediment Rating a;lrve Relationships T- 7 Buena Vista Lagoon Cmputer Model Runs PACZ i 12 12 13 14 16 17 19 23 23 28 37 39 43 - 2 3 5 9 10 26 38 6 18 22 25 25 29 31 35 TA8IrE 8 TABIIF: 9 Sumrrary of IagOOn and "shed Shmlations 36 SUmMlry of Inflaw y.drographs 3 Due to the rapid -tation of Buena Vista Iagmn, the California State Coastal Conservancy funded this study of sedht control for the lagoon and its watershed. Leedshill-Herkenhoff of San Francisco subconsulted and provided the hydrology models of the watershed, several alternatives needed to ke evaluated for sedimnt control. A Phase One study was performed by 3m & Vcqt of Vista. The findings of the Phase One Study indicated that The California State Water Resources Control Board and the Coastal Conservancy then funded a Phase Ttm Study. hired to perform the watershed delling, project coordination, and alternatives evaluation. prepared the hydxaulic analyses of the lagcun. is to formulate a prioritized list of sediment magemat procedures for the Buena Vista watershed and lagcon based on a cost-benefit comparison. June @iega&! & Associates of Carisbad was Philip Williams & Associates of San Francisco The primary goal of this study Buena vista Lagoon is the only freshwater lagoon in southern California. situated between Carlsbad and Oceanside. the cities of Vista, Oceanside and Carlsbad (see Figure I), maintained by a fixed weir at the muth of the lagoon. railroad, Hill Street, Interstate 5 and Jefferson Street cross the lagoon, Since the psition and size of the muth of the lagoon was made permanent and the flcw restricted, there has been an increase in the s-tation rate of the lagcon, Increased, urbanization b~ the watershed further accelerated the sedimntation and necessitated the dredging of a portion of the lagoon, It is Its 19 square mile watershed includes The water level is The fills of the If projected future sedimnt rates materialize, the lifetime of mens Vista Lap could be less than 10 years. T€rE L%Cm SySTEM Bum Vista Lagoon was created by the rapid rise in sea level after the last ice age. In its natural state, it appears that the munt of sedi.”t deposited in the lagcon was balanced by the rate of sea level rise. factors have upset this balance, One is the limitation of the lagoon’s ability to flush sediment to the ocean by the placement of a wir at the muth and road fills that cross and restrict the lagoon, The other is the increase in sediment delivery to the lagoon frm the watershed. The cause for this increase is two-fold. Encroachment upon the floodplains that once accepted sedimnt frm the creek during large stonn flacrs has eliminated mst of this buffer and sediment now flms directly into the lagoon, Before the human-caused disturbances to the watershed, creek, marsh, and lagoon much of the sedimnt fran the watershed was deposited before it reached the lagoon. Then in the 1- reach, a large marsh area spread and filtered the stonn water, trapping mre sediment before it entered the lagoon. Tfm Urbanization has increased flws to the lagoon. sediment was deposited on the broad floodplain in the middle reach, I- w w LL 8 3m The only remnant of this protection system is the area west of South Coast Asphalt and east of the Haymar Street cul-de-sac. This reach of the creek is cnmently absorbing very large quantities of sedinmt and is the last natural buffer to prevent saw of tke s&hent frcm entering the lagcon. functioning prtion of this reach is the thick riparian area just east of the Haymar Street cul4e-sac. The best The other sediment buffer was the marsh above the lagoon. campletely filled. Highwzy 78 aril an bo+& si&s of b%nrce SLreet st left of this marsh. found its way through the tules to the lagmn dropping mst of the finer sediment that was left after the riparian area upstream. This marsh has been The cattails in the channel betwen the shopping center and lbad are dl that are Before the filling of the msh, the water sl0tRd and After changes in the watershed, particularly urbanization in the last two decades, increased peak flm have dramatically transformed the natural hydrological system. the watershed into the creek, but, mre importantly, have caused the creek bed to start to erode (degrade) in portions of creek. The bcrease in stream flow energy has caused this stream dawncutting. new and lower equilibrium can be reached (see Figure 3). The increased flaws not only discharge mre sedinent frcan The process will continue until a All of these factors contributed to the rapid filling of the lagoon from 1978 to 1981, the years that marked the end of a 13-year dq period in this region. In 1981 the eastern portion of the lagoon was dredged. study is to explore alternatives to continued dredging of the lagoon. The purpose of this THE STUDY Hydrolcgy ncdels of the watershed and hydraulic mdels of the lagoon wxe developed and evaluated. in a -11 reduction in the sediment accumulation rate. Greater reduction in the sedht delivery rate can he produced by reducing the peaks of the storm flaws into the lagoon. Reductions in the peak flms *re deled by optimizing feasible detention basins in the upper reaches and by mde1ix-q channel enhancamnt that muld slaw veiocities in the milic;iie reach. Maxuru * 'zing the flushing action of the lagcon resulted Regional sediment transport curves =re integrated with the inflm hydrographs in the hydraulic del of the lagoon. transported is an exponential function of the quantity of water, the peak flows are responsible for a laxge portion of the sedinu3nt transported to the lagoon. Reducing these peak flw resulted in much larger reductions in sediment flaws into the lagan. The study reviewed nine alternative for controlling sedimentation of the lagmn (see Table I). These include: L1) dify the lagoon flushing action by increasing the weir size, the size of the Kill Street Bridge and clearing the opening under 1-5; L2) dredging a flow-through channel through the lagoon from Jefferson to the weir; Wl) construction of stomter detention basins in the upper watershed; Because the quantity of sediment WS) enhancerent of the main channel to lmr main channel 4 0 0 ESTIMATED ESTIMATES) PERCE3ST lwNLTAL COST- INITIAL ANNUAL muc- COST l?mEFIT CCST O&MCCST TION REDX’N 1 RATIO 2 PHASE I1 IAGCCN MIDIFIcpiTIoNS: $15,000 12% $168,000 3.0 ~1 WEIRi-HILL ST+I-5 $570,000 .I L2 CHANNEL $4,030,000 $2,010,000 14% $197,000 PHASE I Wl DGT BASINS Const. $92,000 $139,000 ws CREzKENEmIc $800,000 $160,000 32% $440,000 2.8 WATERSHED mIF1CATIONS: 20% $281,000 2.0 Land Acquisitions $750,000 mIF1CATIoN WITH s” so- a”L: SEDIMENT SOURCE cxxcF.oL: S1 E.C.R. SIDE -YO $30,000 $6,000 1% $7,000 1.2 s2 GRADEDAREAS $6,000 $6,000 3% $37,000 6.3 s3 AGRICULTURAL $6,000 $6,000 2% $33,000 5.6 s4 S.” SED BASIN $12,000 $12,000 1% $18,000 1.6 s5 JEFF. SED” $5,400 $5,400 1% $12,000 2.2 No MODIFICATION: $1,410,000 3 BASIS OF CCME“m 1.0 I. Annual cost reduction is equal to the percent reduction ~~TES $1,410,000 (the estimated annual cost of dredging with no modifications to the lagoon or watershed). Annual. cost reduction is the annual benefit. Estimated total annual cost divide by the annual. benefit. This figure is the annual cost of remving 191,500 cubic yards of estimated sediment delivered into the lagoon at $7.35 a cubic yard for 2. 3. Further details on the derivation of these figures is included in Appendix Two. 6 ss s E c 'r IO N Figure 3 Natural Conditions ........ : .;I .C ......... \.. ..... ......... .... r) 1 1 u v i 11 m ,:.:.:.:.I:: .................... .... .... .>P .................................. ............... ......... -_ --L-.--- 5 . 0 that for every cubic yard of erosion controlled, there is one half of a cubic yard of reduction of sediment accumulation in the lagoon. thirds rduction in the grading site sediment source multiplied by one half indicates that this improveZnent wxld mean a three percent reduction in annual sediment accumulation rate in the lagmn. has the highest rate of return on dollar invested. Similar results are anticipated for agricultural sites (S3) . this program should reduce as the land for agriculture diminishes. reduction of 2 percent of the total sediment accurmlation in the lam per year is anticipated. It is anticipated that 5,000 cubic yards of sedirclent per year dd effectively be remved fran a sediment basin lccated at the South Coast Asphalt property (S4). cubic yards of sedirrent per year, using the source and accumulation relationship esthte described above. in the anrmal sediment accumulation rate in the lagoon. this site wuld reduce the sediment accumulation rate of the lagoon. The anticipated two Table One indicates that this The budget for Currently a It is estimted that this muld result in an annual reduction of 2,500 mving sediment at This dd rean a one percent reduction Another 1,600 cubic yards of sediment per year could effectively be renvwed at the mth of the creek, just east of Jefferson Street (S5) . Since there is no dampening effect expected at this close proximity to the lagoon, it is assumd that this is a direct reduction in the sdimmt acdation in the lagoon, This could result in a reduction in annual sedimnt accumulation in the lagoon of one percent. It is also effective to repair side channel arroyos (Sl) in the manna described for the main channel. !The example given is the channel which is on the west side of El Camino &al (E.C.R.) fKm Hosp to Chestnut. Repairing this arroyo also yields a one percent reduction ad is cost-effective. The alternative that proved to be far too costly was to maintain a deep channel in the lagoon frcsn Jefferson Street to the weir (L2) . It is anticipated that larye amounts of material Fxluld have to be dredged frcan this channel, because the channel wuld act as a sedimnt trap for the lagoon. The high sediment rate of the lagoon nvaans that the channel dd fill in at a rapid rate. It is estimated to cost over two million dollars per year to fiaintain such a channel kcause of the large volumes of material to be rseroved. Refining the estimates with data fran a watershed, creek, and lagoon mnitoring program, and with mre precise cost estimates may or may not improve the cost-effectiveness ratio for this alternative. coNQ;us1ms ~n the order of effectiveness the follawing solutions are reccmekW: creek t=nha,nmt {including floodplain preservation in the laer reach) , detention basins, sediment control education, sediment basins, and side channel arroyo repair. 8 I4 velccities and help repair erosion damage; S2) erosion control education for protection of graded Lands; S3) erosion control education for protection of agricultural lands; S4) a sediment basin at South Coast Asphalt; and S5) a sediment basin at Jefferson Street. alternatives were catpared to no moaifications and the consequential dredging of the lagoon at the anticipated rate of sediment accumulation in the future, SI) side channel arroyo repair; The nine Eight detention basins (Wl) in the upper reaches will produce an anticipated 20 percent reduction in annual sediment: accumulation. of a snall check structure (i.e.: a 5' high dam or a road crossing) and a restricted outlet like an 18" pipe. This allm the water to poM Zing stom and to be released at a greatly reduced rate, basins proposed h this alternative (see Figure 4 for detention basins in Vista). A detention bash consists There are eight detention Creek enhanmt (FJS) proposes peak flm velocities to be less than six feet per second, allcwing riparian grmth in the creek, with a 15 to 30 feet increase in creek width and drop structures which reduce the grade of the channel and thereby reduce the velocity of the water (see Figure 5). Creek enhancerent was delled With the detention basins in place. The &ination of creek enhancement and the detention basins resulted in a 45 percent reduction in annual sediment accurclulation because of the resulting reduction in storm flows. reduction, it is asd that the reduction in storm flaws due to the creek enhancement alone can reduce sediment delivery by a mini" of 25 percent. Reducing the erosion of the main channel (which is the major sediment source in the watershed) by this enhancement will further reduce the sediment accumulation by an additional 7 percent, for a estimated total of 32 percent reduction. preserved , 'This can be accanplished Since the detention del alone yielded a 20 percent Additionally, the 1- middle reach floodplain should be The ambination of a nrrmble and 80' wide weir, a 100' bridge aPening at Hill Street and dredging under 1-5 (Ll) produced an estimated mud. sedirru3nt accmulation reduction of 13 percent. drawbacks that can not he measured by a dollars and cents canprison. the mir will create other problems that are not related to sediment accumulation. Mong those concerns are water quality, and salt water intrusion. levels in the lagoon. warning it and increasing algae grd. probability of ocean waves overtopping the weir, intrcducing salt water to the lagoon. that the channel dredging under 1-5 will have to occur frequently. Hmever, this alternative has major Lawring This alternative increases the probability of lower water surface Imering the weir also increases the Shallow water allows sunlight to penetrate the water, Another anticipated problem With this solution is that it is likely Sp&g a relatively mall annual sum of mney on education for sediment control on grading sites (S2) will produce an estimated tw thirds reduction of this source, Reaucinq sediment sources will change the sedimnt rating curves, thereby resulting in a reduction of sediment accumulation in the lagoon. There is no data on the relationship betseen sediment source reduction and the reduction of sediment accumulation in Buena Vista Lagoon, This study assums 7 .. rl) . ... _. .. . . . . .. -. . . NORTH i *...-. : .' b I j ' CHANNEL LOW FLOW i >- ! -' \ 100-YEAR, 6 FEET PER SECOND CHANNEL *.-- ___ ____ ___ -__-_ -_ r I Figure 5 Cross Section of Proposed Channel 10 LEGEND e1 PHASE 1 DETENTION BASIN FOR STORM ATTENUATION (4 LOC'NS) PHASE 2 DETENTION BASIN FOR STORM ATTENUATION (2 LOC'NS) CHANNEL ENHANCEMENT DOWNSTREAM OF BRENGLE TERRACE PARK TO MODELED IN PHASE 1, BUT NOT PHASE 2 MODELED IN PHASE 2, BUT NOT EFFECTIVE e2 COLLEGE AVENUE. ' Figure 4 STORM ATTENUATION BASIN & CHANNEL ENHANCEMENT MAP - 9 e 0 -. The study found sans lagmn mdifications to be cost-effective, however there are major concerns for the ecology of the lagoon if these are implax?nted. The the Hill Street Bridge to be 100' long, and keeping the opening under 1-5 cost-effective dfications included a "able and widmed weir, enlarging dredged. 11 This is a report of the Phase l3m study of sediment control for 3uena Vista lagoon and watershed for the California Coastdl Conservancy and the State Water Resources Control €!card. several alternatives needed to ke evaluated for sedht control, Applegate and Associates, Civil Ehgineers of Carlsbad was selected by the Coastal Conservancy to perform this study, Francisco was hired to sulxonsult and provide the hydraulic analyses of the lagoon. evaluation rere perfod by June Applegate and Associates an6 are sm".rizd in this reprt, study and has greatly expand& and revised their reccarmendaitons, of in-depth research into the Phase One recohrmendations has resulted in the divergent findings of the Phase Tbm report. Findings of the Phase One Study indicated that June Philip Williams & Associates of San The watershed modelling, project co-ordination, and alternative The Phase M study relied upon the findings of the Phase One This degree It is the primary god of this study to reprt a prioritized list of sedimsnt managerent prccedures for the kena Vista watershed and lagoon based on a cost-benefit camparison. The mens Vista watershed is an ungauged watershed. tkis study are based on synthetic hydrcgraphs, synthetic mathemtical models of "he lagoon, and sedimnt rating curves extracted frm other similar small coastal watersheds, sediment mnitoring using detailed bathmetric surveys of the lagoon after major storm events, stream flcw sediment monitoring, and rain gauge informatian during storm events will yield mre accurate predictions of sediment axumulation, sediment flms and sources, Due to limited budget, ahst no field data, and a range of uncertainties, the sdimnt predictions in this report have a wide range of error changing the true cost-effectiveness analysis dramatically. Consistent techniques were applied uniformly for each categoxy of dification. The estimate of reduction of mu& sediment accumulation in the lagoon provided a relative rankinq of the effectiveness of each alternative. categories of alternatives in which the relative cost-effectiveness ranking is valid is: maximization of the flushing ability of the lagoon with lagoon mcdifications (Ll and L2), minimization of the sediment delivery to the lagoon with mtershed rrcdifications (Wl and WS) Watersm All of the predictions in The and minimization of erosion in the (ws, and s1 through S5). Buena Vista Tagcon was fo& during the rapid rise in sea level at the end of the last ice age, ten to fifteen thousand years age. seven thousand years sea level rise has been mre gradual, apparently abut 1/2 foot per century. natural sedimentation rates in the lagoon, alluwing it to survive for thousands of years. During the last five to The slow rate of rise was sufficient to capensate for Wave action caused the Littoral transport of sand along the coast, sealing off the entrance to the lagoon with a hrrier beach. Esccept possibly during an 12 t -- lagoon. The Watershed has also changed dramatically, farming, removed soil cover, increasing erosion and -tation in the lagoon. increases flood peaks, causing gullying and creating arroyos. greatly accelerates erosion and dcrwnstream sedimentation. floodplain of ~~ena Vista Creek has been filled in several locations, reducing the filtering effect of the riparian vegetation. The floodplain in the middle section of Buena Vista Creek has evolved as a result of sediment kan the surrounding hillslops being transported as bedload in the creek. and deposit sediment on the adjacent flccdplain. built up over time. the middle and bwer reaches in ancient tks (see Figure 3). of the total sediment eroded in the watershed actually reached the lagoon. Extensive grazing and, later, Urbanization, which has been particularly rapid since the 1970~~ This process In addition, the During flood flows, these flows Fxnitd overtop the creek banks The floodplain and the creek Buena Vista creek PELS aggrading in this manner throughout only a fraction After changes in the watershed, particularly in the last tsm decades, the increased peak flows have dramtically transformed the natural hydrological system. The increased flows not only discharye more sediment fran the watershed him the creek, but, mre @ortantly have caused the creek bed to start to de (degrade) in portions of the creek. The increases in stream flm energy have caused the stream to downcut. until a new and lower equilibrium can be reached, This process will continue At the same th one of the largest sediment buffers in the watershed has been filled in. The marsh above the lagoon was the spreading area for storm flaws. Here the water sld and found its way through the tules to the lagon, dropping mch of its sedinaent load. All that is left of that marsh area is a small area where cattails ~rcrw in the channel be- the shopping center and Highway 78. Buena Vista Creek Water flowing into Buena Vista Lagoon comes primarily through mens Vista Creek. covers areas of Oceanside, Carlsbad and Vista. the watershed, covers over half of the watershed and has the major impact on the flow characteristics of the system and yet it does not border the lagoon. The 19 square mile watershed that drains into the creek and lagam Vista is in the upper reach of The creek can be identified by reach. The upper reach is the reach abve Highway 78. reach is frm El Camino Real to Jefferson Street (see Figure 2) . The uppr reach is in its natural channel to Brengle Terrace Park. Above WilM Park the creek has been stabilized by check structures, They were built in the 1930’s and it appears that only one of them has failed, channel 11clw appears to have the potential to overflow and flood the “ding The middle reach is from Highway 78 to El Camino Real. The lower This area, probabli-bause of watershed. The me& bas dawntown area of Vista to increases in runoff fran the urbanization of the been channeled into concrete structures through the Melrose Drive. These structures were also designed 14 - 4 a- early stage in its evolution, the tidal prism of the lagcon was insufficient to scour a channel across the beach. Tidal action occurred for a short period in the lagcon, and only hen winter flocds opened up a channel, Sarne s.edimnt carried into the lagoon during floods discharged directly to the ocean thmuqh the apening. %me sediment deposited in the lagoon muld later be resuspended by wave and tidal action, and muld be flushed out of the lagoon during ebb tide. Because of the varied greatly winters, storm overtopped the reestsblishzd, of Euena Vista barrier beach across its muth, the character of the lagcon fran season to season, and fran year to year. runoff filled the lagoon with freshwater until at same point it barrier beach. In the spring, after the beach was the kgwn level dropx6, fed snly by spricqs and the base flow Creek. In s"er and fall the inflcw decreased further until In normal evaporation exceeded inflow. flats or salt pans mdd be -sed. the lagcon bst cmpletely dried out, and was fed mainly by salt water seeping through the beach, Water salinities increased and large areas of md During drought periods it is likely that In its natural state the watershed of Buena Vista faqmn was covered with native plant vegetation. erosion than many of the introduced, new species. against direct erosion from rain drops but allowed the infiltration of storm moff into the soil, reducing peak runoff rates duwnstream and providing greater base flm in the creek later in the year. These flows supported dense riparian vegetation along the creek banks and on its floodplain. In its lower section the creek muld have discharged into a tule freshwater marsh at the upper end of the lagoon, vegetation acted as sediment traps during high flood flows, building up the alluvial floodplain and reducing the amDunt of sediment discharging to the lagoon. The plants generally provided a higher resistance to They not only protected The floodplain and marsh Historic Changes In the two centuries since Eurcpans settled the area, man has made major modifications to the natural hydrolqic system. The hyraulics of the lagmn have been ccanpletely altered since 1940, when outlet culverts were installed at the muth to regulate maximum water levels, This eliminated uciai action until the big €id of 1963, kkar th ciilvsrts were washed aut, re-creating the natural entrance for a short -&ile mtil the existing fixed outlet weir was installed in 1970. have been excluded and the lagoon has been converted to a freshwater lake, Since that tjm! tidal flows The construction, first of the Hill Street road &ankn?ent and then of the 1-5 freeway e"t across the laqmn, has also affected the hydraulics by segregating the lagoon into three distinct basins. These changes have greatly limited the lagoon's ability to flush sediment out to the ocean. Urbanized runoff pollutants discharged directly into the lagoon degraded water qyality. In addition, until the 1960s, sewage was discharged directly into the 13 0 _- The marsh has been filled. Presently the only evidence of that -&.area is a small area where cattails grow in the channel betwen the shopping center and Highway 78, This channel is nuw the lower reach of Buena Vista Creek. The Iagoon Because of the changes in the lagoon hydraulics and the greatly increased rates of sedimentation the lagoon is no longer in equilibrium with natrual hydrologic processes and is rapidly silting in, Most sediment deposited into the lagoon is discharged during the peak flows of large stom, such as those in 1969, 1978 and 1980, Sediment discharge is exponentially related to the flow velocity by a y of tm to three. Consequently, though peak flows may last only a few hours, they can carry tens of thousands of tons of sediment into the lagccm. It appears that sediment carried into the lagoon is prebmnan ’ tly silt. The typical size distribution of suspended sediment for different .so- is shcwn in Table 2, The Table is based on sediment sampling on other San Diego County coastal streams (see Appenaix 4) and synthetic flood hydrographs generated for the Buena Vista Creek watershed. As the flocdflm approaches the lagoon, backwater reduces the velocity and carrying capacity of the flw. sands, appears to be depsited on the floodplain and in the chamel betvieen the South Coast Asphalt quarry and Jefferson Street, while most of the suspended sedimnt, consisting of sands, silts, and clays, are discharged into the ligoon. As tfie flood flows enter the lagoon, flow velocities drop to a fraction of a foot per second. suspension, and particles start to settle out. The settling velocity of sands if very rapid, so they tend to settle out i.xm&iately. gradually ht rapidly enough for most of thm to be deposited upstream of 1-5. Clays settle out very slowly, but because velocities through the lagoon are so low, mst are depsited in the lagmn and only a fraction are discharged to the ocean. The roadfills of 1-5 and Hill Street and the outlet weir have reduced flow velocities through the lagoon, increasing sedimentation, Lagoon now acts as a very efficient sediment trap. Much of the bed load, consisting of ccarser These velocities are insufficient to keep the sediment in Silts settle cut more mena Vista &sed on the very limited boring information available, it appears that, prior to 1940, the lagoon bed consisted of fine sands at an elemtion of about -1.5 ft NGVD. the lagoon in the vicinity of 1-5, and by 1982 an additional 2.5 ft of organic rich silty clay had accxlrraxlated. In the last 42 years, presuming the same siltation rate over the 200 acre lagoon, this amunts to about one and a half million cubic yards or tons (for these sediments, a cubic yard eighs roughly a ton), or about 35,000 tons/year. For the 19 square mile watershed this amounts to 1840 tondsquare mile/year which is canparable to an earlier estimate of 1,000 to 2,000 tons/square mile/year (Irnnan 1976). By 1961 approximately 2.5 ft of organic rich rnud had accumulated in .. 16 vr a- & built before the major urbanization of the area. The middle reach was a Wide aggrading flocdplain. two by the falls just downstream of College Avenue. functioned very similiarly before the impact of urbanization occurred. the falls mark the location of the hyciraulic division of this reach. It has been divided into Now The tm sections In the upper middle reach, fran Highway 78 to College Avenue, the creek has dramatically changed frcxn trapping sediment in its broad floodplain (aggrading) to a dam-cutting creek. has cut into the ancient flooc3plain deposits in the area of the old sewage treatment plant in Vista. plus the eroded mterial dawnstream. This difference is marked by the large gully which Instead of absorbing much of the sediment it receives as it had done in the past, this reach is nm sending that sediment Currently the lcrwer middle reach (College Avenue esterly to El Camino Real) is still aggrading. feet per season. of the sediment before it gets to the lagcon. In fact, it is aggrading at an alarming rate of up to four This indicates that the later middle reach is absorbing mch The cause of Chm-cutting in the upper middle reach is the increase in the scour action of the creek. road crossings has contributed to this degradation, but this is a long-term phenmon, these crossings with sediment as the middle reach of Ima Alta Creek has. llxeriq the flcw line of the channel at various If the creek still had aggrading characteristics it wuld fill in Typically during urbanization downstream channels start to de. response to higher peak flows of water frcan the developing area. and future hydrographs modeled in the phase one study indicate that the peak transport capacity of the creek mre than doubles. This is in The existing flows have the potential to double in this reach. This mans that the sediment If the sediment transport capacity increases to the pint that it is greater than the sediment entering the reach, the creek then has enough energy to cut into the bed material, when it had too mch sediment to transport it dropped its load on the wide floodplain. these ancient fluvial depsits and carries them and its original load downstream (see Figure 3). Nm it is hungry for sediment and erdes into The increase in pak flow -lains the cause of the erosion in the upper middle reach. the peaks in the lower ~ddle reach to increase dramatically and could trigger dam-cutting in the 1- middle reach. Since there is very little buffer between the her middle reach and the lagoon, the lagoon will expzrience an even greater sedimentation accumulation rate than we have previously seen. It is also a warning. Channelizing the upper middle reach will cause In the lower reach, above the lagwn there was a wide flat marsh. location of El Camino &a,l the creek spread over the marshland. of the land and the thick grcwth of the marsh plants greatly reduced the velocities of the water. tules to the lagoon, dropping much of its sedimnt load. filter for the water entering the lagoon. At the The flatness Here the water slowed and found its way though the The marsh acted as a 15 ._ . .I ... storm - 2-yr 5-yr 10-yr 25-~ 5 0-yr 100-yr TABLE "I. sediment Size Distribution for Different yrdrogr aphs sedim3t peak madat % by Particle Size at Peak Flow Flow Peak Q (cfs) (tons/day) F (-004 .004-.016 .016-.064 .064--25 .25-1.0Qnn 11 9 8 J 1371 1.2~10 25 6 23 33 5 5 1979 3.2~10 23 5 26 38 2589 6.8~10 20 4 28 40 6 3659 1.7~10 18 3 6 5832 6.7~10 15 2 31 44 6 32 47 4 7 I 13 2 33 48 4 7200 1.6~10 18 . .. . .. .. .. .. . r . . . : After initial deposition during and after a flood, sediments can be resuspended by wave action and redistributed in the lagoon, constant, deepening to tw to three feet in areas of hi& wave action. thick growth of tules under the 1-5 bridge prevents all except the finest sediments frm circulating into the vestern segnents of the lagoon. Consequently, sediments acdating in the eastern segment are mainly sands, silts and rmds, wfiereas to the west of 1-5 sediments are minly organic mds. In the wind-protected area between the railroad and the beach, sediments are highly organic and appar to Contain sewage sludge. Water depths are fairly The Water levels in the lagoon are maintained at a mini" elevation of 5.8 ft NGW) by the outlet eir crest or mre cmmonly b-en about 5.8 ft and 6.5 ft NGVD by the barrier beach forming across the ouflet, segment are typically 1.5 to 2 ft, and in the =stern segments 2 to 2.5 ft, S- ??at= dep*&i i? the astern Before the human-caused disturbances to the watershed, creek, rnarsh, and lagoon, much of the s-t fran the watershed was deposited before it reached the lagoon. reach. storm water, trapping mre sediment before it entered the lagoon, sediment was deposited on the broad floodplain in the middle Then, in the 1- reach, a larqe marsh area spread and filtered the The only remnant of this protection system is the area west of South Coast Asphalt and east of the Haymar Street cul-de-sac. This reach of the creek is currently absorbing very large quantities of sediment and is the last natural buffer to prevent same of the sediment frm entering the lagoon. functioning portion of this reach is the thick riparian area east of the Haymar Street cul-de-sac, The best 17 Agriculture The following is a sumnary of interviews with Howard Mueller, who is currently a consultant and has worked for the Soil Conservation Service. for the Soil Conservation Service, he "piled the "Impartant Fannland Map" for While mrking The a"t of fafinland in the mena Vista watershed has been a constant of approximately 5 percent (app"ate1y 600 acres) avocado groves and 5 percent truck crops. farmrs these lands will decrease and becane an insignificant sediment source within 5 to 10 years and will beccane completely urbanized within 20 years. Howlever, with the Various econcdnic and social pressures on the Avocado groves of three years old or younger (of which there is 10 to 15 percent of all groves) produce 15 to 20 tons of sedimnt per acre per year. The rmabhg 85 to 90 percent of the groves have sufficient canopy and leaf litter to reduce the sediment yield to one to two tons of sediment per acre per year- Well managed truck crops also have a sediment yield of one to two tons of sediment per acre per year. Hawser, poorly managed truck crops produce as much as 20 to 30 tons of sedinwt per acre per year. Historically, the poorly managed truck crop land constitutes about 1 percent of the watershed, and the well "ged truck crop land constitutes the remining 4 percent. Presently, the ell managed farms only constitue one and a half percent of the 5 percent total. Historidly, the agricultural land produced an estimated 5800 tans of sediment per year in the Buena Vista watershed. esthated 13,000 tans of sediment per year and within the next five to 10 years Presently they are producing an this source Of sedjmnt will be negligible. By canpadson, the natural areas, ach had periodically burned, yielded. one to tvm tons of sediment per acre per year, average. Future Lad Uses The following estimates for future land uses viere made by each of the three cities in the watershed (in acres): Carlsbad Vista Oceanside Deve~~ area: Underdeveloped: 1700 5760 3000 300 1500 500 Vacarrt land: Area with approved plans: 2 50 2750 134 =ea currently being graded: 190 3 25 85 =ea without approved plans: 250 765 2000 20 channel Erosion It appears that there is an estimated totdl of 173,000 cubic yards of erosion in the min channel of the creek in the and College Boulevard. the spring of 1985. Based on newspaper clippings, most of this erosion has occurred since 1978. 1978 marked the end of a a 13 year dry period in this area. urbanization in that time. The ccmbination of the increase of rainfall and urbanization created a dramatic increase in the munt of runoff ex_periend. What had keen an aggrading section of creek where sediment was deposited upon a broad floodplain becarre a degrading section, gully that is in excess of 15 feet deep in some places in this reach, currentlv Gra6ed Areas middle reach between Mehose Avenue This was calculated by using a real toFography flm in Lard use had changed frm primarily rural and agricultural to The down-cuttinq is evident in a Presently there is approximately 600 acres of graded area in the watershed. Vista, Carlsbad and Cceanside adopted sediment control ordinances as part of their grading ordinances. sediment control plans mst b filed in September and approved by October. The s-t control is supposed to be in place in Novenhr, and remain effective until March, Generally, on-site sedimznt control is still not effective. be a lack of understanding as to what it takes to keep sediment fm leaving the site, 1984 had no protection what so ever. in the other cities was not effective. There appears to Much of the mrk done in the creek in Vista during the winter of Much of the sedkm-k control as installed Sorretimes sediment is directed into the stom drain system in the form of muddy water. just folded aver. before spilling their contents and adding to the sediment leaving frcm the site. Another mn practice is to use sand bags that were not sealed, but Some of these bags do not make it through the Winter season Education of the omtractors, inspectors and design professionals who suhnit and those wfio revim the plans is desperately needed. The ordinances are good, '=It .bo.Iy zre Pit ;et *L?g $21177 1 iEp1me?.?+A. The local citizens are concerned and wuld be excellent watchdcgs, if educated, The Buena Vista Lagcon Foundation has a hotline that is available for the rep0rtb-g of sediment problans. The local citizens are dedicated to the saving of the lagoon, but do not yet recognize some of these problems, Sedimnt yields fran these graded areas can vary fran one cubic yard per acre control is so important. Sediment yields are still high frm this source, so using an estimate of 25 cubic yards per acre per year over the 600 acres currently be+ graded gives a sedkt estimate of 15,000 cubic yards of sediment per year. per year to 30 cubic yards per acre per year. This is why good sediment 1Y . The underdevelopd land is part of the total of the developd area, hmwer it is likely to be redevelo@ to a higher use. Natural Erdible Areas These &rs indicate that there is approximately 6,000 acres of natural erodible area. Using an average of tm tons of sediment per acre per year indicates an estimate of 12,000 tons of sediment fram this source per year. south coast Asphalt South Coast Asphalt has a rock quarry just west of College Avenue. excavatioIls extend approxifiately 50 feet below the streanbed. maintained the falls by keeping the banks and adding levees. portion of the creek, there is runoff fran the plant site and there is potential for erosion. faces do not have mch erosion potential. in ~e reach with loose dirt banks. photography of the lard for this study. Existing channel velocities in all but the mallest flows are in excess of 6 feet per second. been carrying so mch sediment in recent years. continues to aggrade. could start to degrade. In the first large flow in 1979 a portion of a southerly stockpile was ercded. !b prevent this loss of material, South Coast Asphalt nuw protects the tockpiles with riprap at the toe. However this site has the potential for erosion because of above mentioned velocities, little bank protection, and fill-dirt stream banks that are vulnerable to erosion. The In the flatter They have Since the quarry area is granitic rock, the exposed The erosion potential on the site is The plant has sup@ied detailed areal These are erosive velocities, however since the creek has This section of the creek Higher peak flaws could reverse this and this reach The rock quarry is estiroated to close in 1990. the property are now being prepared and reviewed. the creek portion between 1990 and 1995. have a natural looking channel, use the creek as a visual. amenity to the project and maintain the falls. During the time between now and the ultimate developcent, the potential for erosion in this reach can be reduced by widening and adding revetment to the channel. velocities by approximately 20 percent. This will be beneficial in three ways. the sediment transport capabilities of the stream and thereby decrease the potential for creek erosion. reduciq the peak flaws at the lagoon. reduce the potential for erosion. Plans for the developn=nt of The mers plan to develop Ultimately the owners would like to Adding 30 feet to the width of the dirt channel will reduce the bering the velocities will decrease Revetmnt of the creek banks will It will also make a dl contribution to 21 TABLE THREE S" sOuRa3 SUMMARY Wproximate annual average erosion rates frcm each source in cubic yards per year: Main channel erosion: 25 , 000 Grad& areas: 15,000 Agricultural: 13,000 (~UTel'ldy) Natural erodible areas:: 12 , 000 Side chameel erosion: 11,000 22 Detention basin del (Wl.) &cause the future condition wuld have the major long range impact on the lagoon, the future condition was then modeled with eight detention basins. The basin locations had dl been died in the Phase One study. Ttm of than were re-sized based upn mre detailed topgraphic infonnation. The two re-sized basins *re at 13rengle Terrace Park and at Monte Vista School. Detention Basin Locations The detention basins in Vista (shuwn in Figure 4) are located at: 1. 2. 3. 4. 5. Brengle Terrace Park 6. mnte Vista Sc-1 The other Phase (be detention basins are located at: 7. 8. Creek crossing Warmlands Avenue, apprcoCirnately 200 feet Northerly of Elm Drive .Creek Crossing Wannlands Averme, appradmately 200 feet Northerly of Suemark Terrace Creek Crossing Wannlands Avenue, appraxvM ' Sinaoloa Creek Crossing Ste1;We Lane on the north side of Vale Terrace Drive tely 100 feetwrtherly of Calle The canyon north of Mira Costa College in Oceanside The canyon to the east of the extension of Elm Avenue Creek Enhancement -1 (ws) To del the effects of an enhanced channel, with vegetation, the detention basins wre kept in place and the above mentioned channel was widened, The channel in this third mdel had a bottom width.of 15' above Santa Fe Drive and a bottan wid* of 40 feet wide below Santa Fe Drive with side slopes of 2 to 1 and a Manning aoefficielnt of 0.04. The energy slope is abaut one quarter of a percent. This was the ~mst effective i.mprovem=nt to the del. Table 4 gives the sunanary of *t results of these new dels. The 6 feet per second m axi." allowable velocity is attainable with 0.25 to 0.35 percent slope, a M-g coefficient of 0.040, 2:l side slopes and a 40 foot bottan in the upper middle reach reducing in width up the creek until it is about 15 feet wide. Table 4 lists the summary of the peak flows and lag tims at the lagoon for each del. future condition, future condition with detention basins, and future condition with creek enhan-t for a location in the dmtm area of Vista, in the middle reach, and at the lagoon. CcknparisOn hydrographs are shown in Fiwe 6 for a point at the lagcan and a pint at Breeze Hill. Figure 6 shaws graphs for the 2 and 100-year stonns in the 24 The watershed analysis was *fold. at reducixq the peak flows fran stom runoff were analyzed. reduction ms evaluated. Hydrologic watershed dfications aimed Sediment source Reducing peak flows Will reduce the sedinment carrying capacity of the stream. There is an expnential. relationship between the stream's sediment carrying capcity and the water flow. results in a MU& higher reduction in the sediment delivery to the lagoon. Reducing peak flows alm has the side benefit of reducing flooding, areas identified as areas of concern that will be benefited are in the iow lying areas around the lagoon and areas in Vista. Consequently a reduction in the storm peaks Specific Sediment source reduction will reduce the sediment rating curves of a watershed. water flm to sediment transport. reduction for sediment control could be an order of xragnitude off. the cost for greatly innproving the source control is relatively low, retum on the dollar spent is very high. The sedimerit rating cwves are a plot of the relationship of the Hawever, The estimates of sediment accumulation Encouraged by the peak flow reductions deled in the Phase One study, the data fran those original watershed mdels were reentered into the Army Corps of Engineers Hydraulic Enqheerhq Center's H.E.C - I hydrology del and rerun on a mainfrarre ccB[Iputer in order to print hydrcqraphs for the spectrum of six storms. SO-year d the 100-year storms. Those six storms rere run for the Phase One's "ESristhg Condition", '"Future Condition" and "Existing Condition with Detention", Very late in the Phase l'" analysis sam problems in scare of the assumptions for these ircdels were noted and the watershed had to be rdeled. The Phase One input data was used as a skeleton and a new watershed del was created for the Future condition. Since this is an ungauged watershed, the model could not be calibrated. Appendix One describes the calculations for the Future Conditions. developmnt types, Soil Conservation Service Lag, and SCS curve number for eqch af the Phase One shhshs, mre run on this del, Future Condition, Future Condition with Detention and Future Condition with Detention and Channel Enhancement are copyrighted c 1985. Funding was not allocated for this re-mdelling.) These rere the 2-year, the 5-year, the 10-year, the 2!5-year, the Tkese include a suna~lry of the soil types, future The 2-year and the 100-y~~ hypthetical storm (Note: The future condition watershed models for The portion of the upper and the entire upper middle reach) is deled as a trapezoidal channel having a bttan width of 20 feet, side slopes of 1.5 to 1 and a Pw coefficient of 0.02. that is king placed at Breeze Hill. channel beginning below Brengle Terrace Park to College Avenue (a This is typical of the Tri-lock,chamel 23 sm of H.E.c.- 1 pak flows and time to peak for future conditions at the lagoon. 100 year: Future oonditim 13,734 cfs 2.94 hours With detmtion 11,431 cfs 2.95 hours With enhancenent 9,231 cfs 3.28 hours 2 year: Future condition 3,213 cfs 3.27 hours with detention 2,590 cfs 3.. 28 hours With enhancmt 2,130 cfs 3.82 hours NO E’UTURE EXISTING BASIS L1 ?.?“a CDMB 12% ET“E cm” 26% ~2 EVNRE CHAN ~IBVTION 14% W1 F W/DE;T EXISTING 20% DET+ENH EXIT 52% ** ws E”c MIST 3 2% ** 45% FMIcfi Is DUE To STORM EWW ATTENUATICH PLUS 6.5% REDUCTICIN IN ANNUAL SEDIMENT mTIcR3 DUE To THE ”Im QF THIS SED= ExxEm 25 15,000 cFs I LAGOON CFS 3,000 2 3 4 5 HOURS 10,000 5,000 0 2 3 4 5 HOURS 100-YEAR STORM 2,000 1,000 0 CFS \ BREEZE HILL HOURS 2-YEAR STORFi c 5 4 ‘3 IiOURS 100-YEAR STORM 2-YEAR STORK ABBKEVIATICN FC FUTURE CHANNELIZED CONDITIW FD FUTURE DETEI!TIOII KOOEL FE FUTURE WITH DETENTIGN AIJD ENliAI,ICEIiED CHAKFIEL EODEL Fi ljure 6 Hydrographs 26 To detedne the mst effective means to reduce sediment accumulation in the lagoon, its sedhznt budget rnust be estimated under different conditions. A sedimtnt budget is simply an accounting of th inflaw, outflow and stroage of sedimnt in the lagoon for a particular tim period. Sediment inflow is determined by empirical relationships betwsn the variation of sedhent discharye and flow rate during flood events. The sequence of flow rates, &wn as a hydrograph, are ocanputed for particular stoms using a standard cqmtw del simlati. referred ~ as HtE-I, When the sediment enters the lagoon, som settles aut, some is discharged to the ocean, d scm ramins in suspension for a particular tinre period. The anrmnt settling aut is calculated using settling velocity relationships for each particle size. The imrsunt remaining is suspension is detemined by the difference in sdimmt inflw and the a"t settling aut in a particular the period. The munt discharged to the ocean is the sediment concentration remahing in suspension at the outlet, dtiplied by the discharge volume. Yist se is carried into the lagcon by a few large, infrequent floods, Consequently, the sediment budgets mt be averaged statistically over a long pericd of time to estimate an average annual edkmt budget, annual. sedhent budget can be calculated for different inflcrw and lagoon conditions to provide a ccgnparison of the average annual sedimnt accumulation in the lagoon under different conditions. The average Because of the eno~us ntrmber of calculations required to estimate average d sedinaent accumulation, it is best done using a ccrmputer del that simulates the "ent of both water and sediment through the lagoon system. Such a ccanputer del was developed specifically for this study and is described in succeediq sections. It should be noted that there are a great many uncertainties in mst of the calculations involved in detemimhg a sedimnt budget. budgets of this type auld be used for amparatin purposes only. Consequmtly sedirrrt3nt sediment Inflow The best Illethod for de- sediment discharge to the lagoon is to develop a sedimnt rating curve for Buena Vista Creek. A sediment rating curve plots suspended sediment against flm discharge measured at a particular point on the stream, for mens Vista Creek. based on sample data froan other streams. Unfortunately, neither suspended sediment nor discharge data exists Therefore, a sediment rating curve was constructed, watersheds greater than than those with dler dler watersheds =re Strean gauge data fran 11 gauging stations on coastal streams south of Dana Point bere exarmned ' (see Appendix 4). The results shawed that streams with 100 square miles had different sediment rating curves watersheds. consequently only data frcm the five used. 28 Sediment Source Control M? Channel (W): The average of 25,000 cubic yards of sediment channel primarily occurred duriq the 5 years source will not k “pletelv eliminated with per year ercded frm the min frm 1978 to 1983, the channel rtdifications, the Althouqh this reach of the recc”d&i &cerrcent is at least three times the length of the kadly degraded section frm where the 25,000 cubic yards cam. Protecting the Main channel will reduce the sedinaent source to the lagoon by 25,000 cubic yards per year, to the reduction in sedinaent delivery for the “Channel Enhancement” alternative, because it is assumed that for every cubic yard sediment sou~ce reduction there is a eorpespoilding rductio~ of a hzlf of a cubic yard of reduction in the sediment accumulation in the lagoon, For the camparative analysis half of this quantity was added Graded Areas (S2): The adopted sediment control ordinances for graded areas are good and mney is being expended by the developers and the cities for the design construction and review of sediment control plans, However, there is still an estimated 15,000 cubic yards of sediment escaping frm fran these sites. sand and gravel kgs are allowing sedimxk fluw into the storm drains and channels, surprise rainstorm, Misplaced or broken Smtimes the projects are caught with their gravel bags down by a Education of the engineers, inspectors and contractors involved will probably reduce this source to one third of the volute it is today, Agricultural (S3): A similar &cation program for agriculture could be as effective as education for grading. This would reduce the sediment source by an additional 5 percent, Natural. erodible areas: This surce has such a lm concentration very little improvement can be mde. Side Channel Erosion (Sl): Three side channels that need reconstruction and enhancement have been identified. Camino Real between Elm and Chestnut and next to &”e, south of Mamon, They are just west of Pamelo Drive, on the westerly side of El According to the Vista City staff and the Carlsbad City Staff the side channel arroyos at Pamelo and Monroe are in areas that are apprcvd f~r dei,*&qnezt, The requirements for the developznts include repairing and premtincj these arroyos. The side channel arroyo on the vesterly side of El Camino Real beheen Eh and Chestnut locations muld effectively be repaired using drup structures in the same mer as the recamended min channel repair in the upper middle reach. This will reduce the total watershed s&imnt source by an estimated one percent. 27 accumulation further constricts outflows under the 1-5 bridge to approximately tke existing lagoon level. The central cell discharges to the western cell through a constricted culvert under Hill Street, or over the top of the roadway, if the water level rises high enough. The *stern cell discharges to the ocean over the existing sharp crested weir. The railway bridge crossing does not significantly constrict flood fluws the. bestern cell, A sketch of the lagcon is Shawn h Figure 7. A flood muting is a sequential calculation that CdLCulates outflow, water surface elevation, change in storage and average flow velocity for a given inflow and initial lag- oonditions. bathymetxy of Buem Vista Lagoon exists. elevation/storage relationship for each of the cells is estimated based on few pint soundlll * gs and available topogfrraphic survey. 4. Outflow frm each cell is desed by standard wir flcw or culvert flow formulae. the average cross-sectioned area for a given instant in time. Unfortunately, no detailed survey of the These are shown in Pspenaix Therefore, the water Averaqe velocities are simply calculated as the outflm divided by Sediment inflm to the lagoon is asd to be unifody vertically mixed in the Elm. When it reaches the lagoon, flow velocities drop considerably and sedbwnt particles settle out. The rate at which they settle aut is &teed by the particle settling velocity. For the median dianeter of each of the five size fractions, this value is ShaJn in Table W. It can be seen that there is roughly an order of magnitude difference betseen the settling rates of each size fraction. can therefore intrcduce another source of error in the sediment budget. Dividing the sediment inflow into five discrete size fractions sediment accumulation is detexrtined by the fraction of sediment that settles out while the sedimnt-laden flaw travels fran the influw to the outflck~ each cell, Hill Street culvert ccanpletely mixes the sediment, so that sediment discharged The -el for sediment depsition descrikd above is strictly valid only for quiescent unifonn flows in stilling basins. entering the lagoon will be highly turbulent, which tends to hinder settling of sedimrtnt particles until the turbulence dies out. Unfortunately there is no feasible way to model this process, and calculating deposition based strictly on tjm of travel. may overestimate sediment accumulation particularly for the finer sediment fkactions. of It is assumed that flow travelling through the 1-5 bridge and the to the nexk cell is uniformly mixed. lh Bue~ Vista Lagcon flocd flows The velocity of the flow in the lagoon itself can generate turbulence that can keep particles suspended, There is this 1- limit of sediment concentration in the lagoon for a particular average lagoon velocity, alternative methods of calculating this lower limit; and as is cc" in the field of sediment hydraulics, there are significant differences in different estimates, For these calculations the relationship developed by Bagnold was used (Yalin 1971), There are ~~ny 30 TABLE SIX. SEDIMENT RATING CURVE mTIcNms USED IN ANALYSIS Sediment Particle Size gs (tons/day) Fa1 Velocity (ft/sec) -5 -4 2.4 Clay (-004 mm 9x10 Q 1.31 x 10 -4 -3 -3 2 -5 3 Silts .004 - .016" 4x10 Q 1.15 x 10 .016 - ,064 1x10 Q 2.3 x 10 -2 -1 -5 3 -3 2.1 1.6 x 10 Q 3.28 x 10 Sar!!S .064 - .25 IIP~ .25 - 1.10 IIP~ 3.7 x 10 Q 2.3 x 10 The suspended sedkt data was broken down into five different size fractions representing nuds, fine silt, coarse silt, fine sand and coarse sand. plots are sham in Appendix 4. data. curves. as the straight line relationships shows in Table 6. calculating sediment inflow to the lagoon. The There is a large of scatter in the Vp to an order of magnitude of scatter is typical of sediment rating However, the rating curves for each size fraction can be simplified These are used in A portion of the coarser sediment discharge, often estimated to be an additional 20%, is carried along the channel bed as "bedload". This was not included in the sediment inflcw because mst of the coarser material appears to be deposited in the Euena Vista Channel upstream of the lagoon, and because it represents only a snall portion of the total sediment load discharged to the lagoon. The sediment inflow for a particular storm is calculated by integrating the sediment rating curve with the inflow hydrcgraph for successive time increnents . Lagoon qdriulics In order to calculate the sediment accululation in the lagoon for a particular flood, a flood routing calculation must be carried out to determine the water surface elevations, volumes and flow velocities at different times, as the flocd flows enter the lagoon. For Euena Vista Lagoon this is ccanplicated, because hydraulically the lagoon acts as three distinct cells. inflow pint at Jefferson to the 1-5 embankment. outflm to the central cell that extends frm 1-5 to Hill Street, The eastern cell includes the area fran the The ak"ent constricts Sediment 29 -_ TABLE7 . BUENA VISTA LAGOON COMPUTER MODEL RUNS MIN NO. INFLOY HYDRUCRAPH UCOON CONDITIONS SEIJlMENT ACCUIIUWTLON RESULTS - In1 ti a1 Di 6- Hi 11 vacer Inflow Deposited charged Corcnts 1-5 Sc. Outlet Weir surface cons Lon6 x.7 r -- - Frequency Watershed Yelr Width invert length clev XCVU Conditions Elev fr culvert ft ft ft 90 IO hrrler weir at original Iengrh:hfrting lagoon ~014. 1enghC:ExiStfng Inyoon evld. 89 11 (rutlet weir ae ociiinil * I 2-yr Existing 2 5-yr Existing 3 257r Existing 4 1007r ~~i~ting 5 Z7r Future 6 5-yr Future i 25yr Future . 8 10lhyr Future 9 .27r Future 10 S-yr Future 11 2s-yr Future I2 1-r Future 13 27r Future 14 5-yr Future 15 25-yr Future 16 lOO-yr Future 17 27r Future 18 3-yr Future 19 257r Future 20 1-r Future 21 Z7r Future 24 100-yr Future 29 27r Future 10071 Future ft elev ft 5.8 15 2.0 5.8 1s 2.0 S.8 15 2.0 5.8 15 2.0 5.8 15 2.0 5.R IS 2.0 5.8 IS 2.0 5.8 15 2.0 1.5 1s 2.0 1.5 15 2.0 1.5 15 2.0 1.5 1s 2.0 5.8 loo 2.0 5.8 100 2.0 5.0 100 2.0 5.11 100 2.0 9.8 15 2.0 5.8 13 2.0 S.8 15 2.0 5.8 1s 2.0 1.5 100 2.0 1.5 100 2.u 5.8 15 2.0 5.8 I5 2.0 80 5.u 5.8 8u 5.8 5.8 80 5.8 3.8 80 3.8 S.8 80 5.8 5.8 ao 5.8 5.8 80 5.8 5.8 80 3.8 5.8 80 5.8 5.8 80 5.8 5.8 80 5.8 5.8 80 5.8 5.8 50 5.8 5.8 M 5.8 5.6 SO 5.8 5.8 50 5.8 5.8 a0 1.8 5.8 80 1.8 5.8 80 1.8 5.8 80 1.8 5.8 80 1.u 5.11 8U 1.8 5.8 50 5.8 5.8 50 5.8 5.8 10.915 26.5110 112.770 872.b40 20,050 51.b35 235,560 I, 147.620 20.185 S1.YZO 235 .%0 1.195.070 2U.OIS 51.165 - 1.143.265 20,060 51 ,6&0 235,965 1,145,875 20 ,060 1.1C7.030 20,075 1.1L9,200 31 10,095 Z4,IZS 1261510 767 .SO5 18.370 46.635 2oa,ns I .011.46S 18,510 4b.985 208,595 I ,012.wo 19.240 46.510 - I.Ol2.31S 17.545 44.945 204.525 994 * 9Y 5 IS ,6bS Y63.440 18,275 1,021.505 89 11 bfrttng wtershed b lagoon conbittons; outlet ueir 80' 88 12 &xisting watershed 6 lagonn conditions; outlet weir-W' 92 8 Cxistlng lagoon b fut.vrter- shed cond.; outlet weir-80' 90 10 Exlrtlng lagoon b fut.Water shed cond.: ourlet weir-Uc,' YZ 8 1-5 mi? lorered; no otki lagoon changes 90 10 1-5 ueir lovered; no ocher lagoon changes 89 11 1-2 weir lowered; no ocher lagoon changer 85 If 1-5 wlr lowered, no other lag- changes 66 14 Hill St. opened: outlet weir at SO* 91 9 Hill St. opened: outlet ueir at 50' Mll St. opened; outlet weir at 50' --- 88 12 Hill SC. opened; outlet uelr ac SO' 87 13 Outlet weir lowered to 1.8' NCVD 87 13 O~tlet uelr lacred CO 1.8' WGVD 87 13 Outlet vcir lowered to 1.8' NCVO U7 13 Outlet weir lwered to 1.8' NGVU 711 22 Combination run: 1-5, Hill St. L 1-5 opened up L(1 16 Coibtnatiat run: 1-5, Hill SI., b 1-5 opened up TABLE 7, Run NO. IN- MOROCMPN UCOOM CoMOlTlONS SIDIMEHT ACCUWLATIOW RESULTS Inltlal 01.- C-ntt n111 uater lnflou Dapsi ttd chbrgtd St. Outlet Ycir aurEace tons toils rt 1-5 -- - frrgu*ncy UatorDhd Mlr Width lnwrt hnsth .lev WVD condiclona clev fc eulmrt tc tt ft tt el.. IC 31 1007r future 1.5 32 IW7r Fucurr I .5 -. 33 12-yr Future I .5 34 27r futum 1.5 3) 100-yr Exlsc. r/dcr. 1.5 36 100-7r Future 1.5 a37 lW7r ExIst.u/det. 5.8 38 LOO-yr Future I .5 39 1007r Tuture 1.5 0 LO 100-w Future -0 -3 100 100 lVU I ou 200 IO0 15 100 IO0 LOO -8.3 -8.5 5.8 5.8 5.8 5-8 5.8 5.8 5.8 100 100 IS 13 15 15 la 15 13 2.1) I .8 I .a 1.7 1.7 1.7 2 .o 1 .8 I .a -8.0 8.0 -8.0 -8.0 2 -0 2.0 2.0 2.0 2 .o 2.0 2.0 nil no 8ci 80 80 80 50 80 80 80 uu IU 80 50 M 50 50 M 53 w 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 5.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 l.n L.6 1.0 1.8 1.8 5.8 5.U 5.8 5.8 s.u 5.1) 5.11 5.0 9.8 3-8 5.3 5.8 5.8 s.8 I, ICY ,310 - 20 .Ob0 34b,135 1,143,470 347,265 1,149,310 I, 143 ,b70 1.141.985 20.1115 >41),2W Y ,JW 9,325 61.636 2.51b,77Y 48.565 .. 1.901.OU2 33.92U 1.369 .Y IC 1.1 90,359 971 ,bkO - 16.U10 282,YZS 965.935 3UJ.?IO 9b8.550 909.160 892.940 Ib.215 tbJ.545 6.360 8,6M 56,592 Z.ZZ5.230 43,YLIl 1.672.570 31,ZW - - - Comblnation run: lntrlat USEL- 1. n * : FATAL ERROR ->)OK of ne*. no. ti5 15 Lontred HI11 St. rutvrri -- 80 20 82 18 86 16 87 13 83 17 80 20 18 22 71 29 7b 21 68 32 93 7 90 10 us 12 91 9 88 I2 YZ a 87 13 Invert to avoid above crr. NO flar out of Basin z or 3 Note, program dld run Loucrad Htll St. to 1.7. 18 avert above problem CP Run t32 b 30 (for ex1 sclng lagoon condl t l on4 Channel birulation: uidtii B rrer-l/IO original con1 tptvh. tlun v-IUX Exircing lagoon condl t1on.: Sar as t3e but depth lncrtsaod by 5 ft IO' channel. SU'uldc, nrc.1 ecalcd dwn by racio ot uidthD (gee SLU.DIT file! lo' deep channel. 50'ulAc: thl8 in kat cane acetimi i. IO' deep channel. IO' ulh this 1. best cmae scrnari vucutv Iaswn rith ohrnnrl Future lrRoM u/ channel Existing lagoon COndfLiUl1. Lcvimed hydrology Rev1 ~cd hydrology Ltvtwd hydrolosf Rcvf md hydrology Ilrvl8ed hydrology Revtsed hydrolony qs is sediment discharye p is the fluid density V* is the shear velocity IBn is the man velocity w is the particle fall velocity h is the depth yf$ is *e kinemtic viscosity This lower limit is particularly important in estimating net accundation of the finer sediment fractions. The total sediment accurrrulation for a particular flood hydrograph for each cell can ke estimated by surrsning the accumulation in each time period. The (2on-p uter plbdel A ccanputer del was developed to simulate sediment acwation in the three cells of Buena Vista Lagcon for a given inflaw flood hydrograph. This model, identified as MPOND c FWA 1985, was an adaption of an earlier lagoon flood rout- model with the additim of sediment routhg subroutines. The input data required is the inflow flood hydrograph, sedhent rating cu~ves for each sediment size selected, lagcon geometry, and weir and culvert-. characteristics. accumulation, and sedimnt discharge, for each fraction for each basin in the lagoon. The ccanputer mdel also calculates inflow, outflaw, wter surface elevation ad velocities of each basin for each hour. hydrograph pericd the -1 calculates the total sediment accumulation and discharge for each basin. The output is a amputation for each hour of the sediment inflaw, sfament Over the entire A total. of 50 ccmputer runs were ded aut to examine the dinations of different scenarios of watershed conditions and lagoon hydraulics. sx"ry of these ~uns is shown in Table Seven. Forty-four of the runs ere carried out based on FEZ-1 inflow hydrographs representing three different watershed conditions that were develop& in the Phase One study. A cqletz These wsre: 33 Y3 1, misting conditions 2. Future conditions with no action 3. misting conditions with detention basins. For each condition the 2-year and 100-year flood hydrograph was analyzd, and in sane instances the 5- and 25-year hydrograhs as tell, errors tere found in these HEC-1 hflm hydrographs. recalculated and six additional computer runs based on three new sets of hydmqraphs were carried out: Late in the study, Accordhgly they w=re 4. Recalculated future conditions with no action 5. Recalculated future Conditions with detention basins 6. Recalculated future conditions With detention basins and preserving the floodplain upstream, A surc~nary of the influw hydrographs used is qiven in Table Eight. Although the earlier HEC-1 hydrographs wxe not accurate, the ccmputer modelling results can still be used as a basis for comparing the effectiveness of different dfiations to the hydraulics of the lagoon, lagoon hydraulics conditions were examined, either individually or in dination: Five different 1. Ekisting conditions 2. Remmhg the sediment accumulated under the 1-5 bridge 3. Enlaxying the Kill Street culvert 4. increasing the capacity of the outlet wir by widening and lowering its crest 5. Esrcavating a deep channel through the lagoon frm Jefferson Street to the outlet weir. The average annual sediment accunilation for selected scenarios is shown in Table Nine. probability curve usbq the 2-year and IOO-year amputer run results for total sediment accumulation and then calculating the average annual sedinvtnt adation by the mthcd described in the ASCE Sedimentation Engineering Handkook Table 4.18. These values are detennined by plotthj a sediment accumulation Table Nine shows that average annual sediment accumulation in about 33,000 tons/year under existing conditions (scenario I). 20 acre feet per year which, averaged over the whole lagan, is abaut 0-1 ft/year. At this rate the lagoan wuld fill h ccmpletely in twenty tu thb2y years. This am~unts to about sediment accumulation of 33,000 tons/year, assuming a trap efficiency 34 -- Source Fld Flood Watershed condition ~ -~ Return Peak Flow vo3.um cfs acre ft. Frequency Phase I study Phase I study Phase I study Phase I study phase I study Phase I study Phase I study Phase I study Phase 1 study Phase I study Phase I1 study Phase I1 study Phase I1 study Phase I1 study phase I1 study Phase I1 study Existing Existing Existing Existing Future Future Future Future Existing w/det Existing w/det Future Future Future w/det Future w/det Future w/det & fp Future w/det & fp 2 5 25 100 2 5 25 100 2 100 2 100 2 100 2 100 1371 1979 3659 7200 1712 2485 4412 8000 1227 5044 3213 13734 2590 11431 2130 9231 429 608 1067 1942 539 741 1245 2145 414 1553 428 1583 - 1483 314 1351 35 TABLE "E, LbGcoN AND mmm s="s Avg. Annual Simulation Watershed Hydrograph Iagmn ccarrputer Sed, Accu. 8 source cod. Runs (tons) reduct. No. Coditions 32,970 N/A 1 Existing Phase I Existing 1,2,3,4 2 Future Phase I Existing 29,30 73,550 0 3 Phase I Future 13 Future CQnb, 32,34 64,230 4 mture Phase I Future Ccanb, w/channel 40,41 56,280 23 5 65 Existing wldet Phase I misting 37,44 25,490 6, Existing wldet Phase I Future Ccsnb, w/channel 42,43 19 , 840 73 Phase I1 Existing 45,46 191,500 0 7 Future 8 20 mture w.det Phase I1 Existing 47,48 153,600 w det of fldplain Phase I1 Existing 49,50 105,100 45 36 U I- D a- 0 0 3 of about 903, corresponds to a watershed sedhxmt yield of 1500 tons/squaremile/year, a value fairly typical of this area, closely to the observed historic rate of accumulation of approximately 35,000 tons/year. Haever, such a correspnclence should be regarded as fortuitous because of the large potential errors inherent in analyses of this type. It also corresponds Sedimmt accumulation under future watershed conditions with increased urbanization and no sediment control is projected to increase by a mltiple of six to about 200,000 tons/year (scenario 7). inpeak flaJs frcan storm drains, lined channels and paved surfaces, and the expnential relationship betrzjeen s-t delivery and flcw rate. The resulting predicted watershed sedhent yield of 7,500 tons/square mile/ye.a- is not unmn in urbanizing watersheds of this type. 200,000 tons/year if distributed equally in the lagoon, could fiii it ip1 €ow to five years. This is due to the increase A sediment accumulation of The hture of the Lagmn If no action is taken and the existing sedimentation rates continue, most of the lag-oon will probably fill with s-t within the next 20 to 40 yeass. Hwver, because of additional urbanization in the watershed (underway or planned), and the filling of the Buena Vista Creek floodplain, mst of the lagoon might fill in in the next 10 to 20 years. ccm gradually butmre likely as a result of 2 or 3 major floods. Consequently, the rate of filling will depend on the sequence of wet and dry years. Tkis sedimntation wuld not There are two strategies for ducing sedimentation in the lagoon: reduce s-t inflow or inprove the flushing ability of the lawn, Table One shows the relative successes of these tsm strategies. alternative for mitigating increased sedimnt flows to the lagoon is channel enhancement (Alternative W. Channel enhancement was rdeled frm downstream of Brengle Terrace Park to South Coast Asphalt, Vaxious channel configurations will accxxnplish this goal but a maxi" velocity of 6 feet per second xlust be maintained, check structures along the cr& and recontour the channel as necessary, This velocity will dlcw for the growth of riparian vegetation; haever, in many areas the existing "natural" channel will have to be redesigned, drop structures installed, and creek banks revegetated to meet this velocity goal. Fortunately ve have a mdel to follow. Gully restoration was acccmplished in the Tahoe Basin using check structures, These have been in place for several years. Not only does channel enhancement decrease peak flood flaws to the lagoon mre effectively it also decreases the erosion hazard for the major source of sediment to the lagoon, either The mst effective The key to keep- the velocities at this level will be to install Mmst as effective in decreasing peak flocd flows is the design, installation and maintenance of eight storm attenuation basins (Alternative WI). The location of these and the channel enhancement are shown on Figure 4. the attenuation locations were mdeled in the Phase One study. Tt;o ere rdeled, based on mre Zetailed toposraphy. .sP1 of 37 The effectiveness They only achieve seen by ccmparing of individual. mdifications in the lagam is negligible, the 13% reduction when implmted together, runs 9 and 12, 13 and 16, 17 and 20, with 29 and 30, and with This can be 32 and 34, as shown in Table Seven, In its present state, lMnaged as a fresh water lake, %em Vista Lagoon is a very efficient sediment trap, it, sedinrent fran the lagoon is to restore it to tidal action. However, because the sediment dischaye to the lagoon is now so large, even with tidal action the lagoon dd probably silt up to a fraction of its original size, with the best pssible scenario of watershed impruveinents, considerable amounts of SediImk will have to be renrrwed in order to maintain the lagoon in its existing state, To pdct the amount of dredging required, flaw msurenmts and at samples will have to ke taken on mens vista Creek and at the outflow weir in order to calculate a mre accurate sedi"t fiudget. bathymetric surveys will need to be made of the lagoon to msure actual sediment accumulation, watersheds, the average dredging requirement probably will be in the range of 10,000 to 100,000 trms/year, assuming the watershed dfications are implemxked, of suitable size for beach replenisfrment, It traps about 90% of all sediment which enters The only hydraulically feasible way of greatly increasing the flushing of Even In addition, periodic Based on the existing analysis using data frm other probably betseen five and twenty percent of this sediment will be VI, 1, - 2, The Taqoon's mture In its existing state Buena Vista Lagcon acts as a very effective sedimnt trap. lagoon will silt in over the next benty or thirty PmSr with mst of the siltation occurrkg during a few large storms. Under future watershed conditions the rate of siltation will increase substantially, reducing the expected lifetime of the lagoon to less than ten years, effective means of reducing sediment accumulation in the lag- is to reduoe sediment inflaw fran the watershed. laqoon can be reduced by about 50% by reducing flood peaks in the Under existing watershed conditions it appears that the entire The most sedinaent acnmdation in the watershed and enhancing the creek, thereby reducing SediREnt inflow, To reduce sediment entering the lagoon fran creekbed erosion, a "m channel velocity criteria of six feet per second is required, effective and was modelled as the maximum velocity that vegetation could conceivably withstand, more effective, This is her velocities are recQmranded and will be All initial creek design should check flood control requirements, and utilize the "I velocities using Manning roughness coefficients of 0.030 to 0.050, which is the range of the vegetation growth roughness in the channel, So the velocity of the 100 year storm should not exceed six 40 It is evident that there is a ptential for decreasing lagoon sedimentation by about 453 by reducing flccd peaks under future conditions by constructing detention basins and preserving and enhancing the floodplain upstream (Alternatives W and WS) (. The third most effective mthod of reducing sediment to the lagmn is stopping the soil loss frcm grading and agricultural operations, for the three cities adequately require sedinent control for grading operations. are effective to amply with the ordinances. Agriculture is still a major soure of sdmnt and there appears to be no ordinances for this source. Agriculture will probably dwindle to be very snall within 20 years, as the watershed is urbanized. The grading ordinances An education process still needs to occur for the methcds that This study envisions a program of education for the City personnel who check grading plans and the operators and building inspectors who are responsible for Lhe outcam of these plans. requirements of the erosion control ordinance and specific recomnendations for cost-effective design and implemntation of the ordinance. me education prcgram muld give a review of the Similiarly the agricultural education program Fxnild focus on agricultural operators and the Soil Conservation Service and emphasize cost-effective implemntation of best management practices in the watershed. Small mmts of the total sedimentcanberemovedinm sediment basins lccated at South Coast Asphalt ard the muth of the lagcon (Alternatives S4 and S5). the material could probably be used on the site. sediment basins is that it is less costly to remve the material at these locations than after it reaches the lawn. The nature of the material remJved will be sandier than the total sedim;nt acdating in the lagoon, therefore a mm usable mterial, The one at South Coast Asphalt will require less costly mintenance, as The advantage to these The sdimmt frm side channel arroyos can be effectively controlled by repairing the arroyo and preventing further erosion (Alternatims Sl). These would be repired in the same fashion as the main creek channel. Sediment inflaw can ke further reduced by control of sediment sources in the watershed, easily quantifiable. Hcwever, it is not unreasonable to suqgest that vigorous application of such "r es CMitd reduce the sediment load by about 50%- This, cmbined with reduction of the El& peaks, muld mean that the sediment inflm wuld be reduced to about 30% of its expected future value. The releative success of these sediment control masures is not In contrast, ncdifying the lagcon hydraulics is not nearly as effective in rducinq sediment accumulation. - clear- the 1-5 bridge, wid- the Hill Street culvert and lmerinq and widening the outlet eir, the sedimnt accurrrulation reduces only by 13%. Even with these modifications, maintaining a deep channel througfi the entire lagoon reduces sediment accumulation by 23% (Alternative L2). require constant maintenance as it Would tend to fill with sedimnt resuspended by mve action in the summer, and is not a very practical option. With all the reasonably feasible dficaitons A deep channel would 39 . The fact that this is a ccqletely ungauged watershed gives the findings of this study an mt of uncertainty. mre accurate estimates of flood flaws, sediment transport and sediment accumulation rates can be observed. The mdels used in this study then can be calibrated and the accuracy inprd greatly. Rain gauge and stream flcrw mnitoring are needed to calibrate watershed Wls. regarding stonn flow values. There is a need for dtoring so These will also help to answex many of the local questions Present estimates of these stom flow values vary wildly. Sediment sampling in the creek Will provide more definite correlation be- the sediment flaws and the flood flows. This will help develop a sedimnt rating curve for the Buena Vista watershed. Bathymetric surveys of the lagoon need to be taken, especially before and after major stom events in order to monitor the sediment accumulation . 42 3. 4. 5. 6. 7. feet per second using a Manning roughness coefficient of 0.030 and the finished floor elevations should be one foot higher than the anticipated water surface elevation using a "ning coefficient of 0.050 for 100-year storms. A low flow channel should be designed to accorrmodate the 2-year stom flaws, possible. velocities and banks revegetated to prevent erosion. cross section is shown in Figure 5. This lm flaw channel should be configured to be in shade wherever Drop structures should be placed in the channel to lower A typical stream The lower middle reach is absorbing large quantities of sediment and is presently offering a significant buffer for the lagoon by acccxMdating great quantities of sediment being transporteti in the creek, of the fldplain must ke preserved in its present state. This section Detention Basins (Wl) The eight detention basins outlined on Fipe 4 should be built and maintained. be the design criteria for these basins. Maximum peak attenuation for the 2 to 100-year stonns should Lagcon Mfications (L1) The mst effective mthcd of decreasing the sediment accumulation in the lagoon with lagoon enhancement is a ccanbination of a mwhle crest teFr with a width to 80 feet, enlarging the opening at Hill Street to 100 feet wide, excavating the material under the freeway, and breaching the barrier beach during storms. sediment accunrulation by only 13% and has many enviromtal drawbacks. This scenario of improvements reduces Continuing Erosion Control Elfucation (S2 and S3) The fixst erosion control wrkshop was ell attended. ccaranunity aCanmitnEnt to erosion control. both for grading and agriculture will pay off well in terms of reducfns sediment accumulation in the lagoon be beneficial. There seems to be a Efficient erosion control sediment Basins (S4 and S5) The txm feasible locations for sedimnt basins are at South Coast Asphalt and just upstream of Jefferson Street. Drdgbg Even with all of have to continue compared to this the effective mthcds outlined herein, the lagoon will to be dredged periodically. The alternatives were all inevitable and ongoing solution. 41 .. ASCE "ais and Re” on Ehgineerhg Practice, No. 54, 1975. sedimentation Hofhan, J. S., et d. 1983. Projecting Future Seal. &vel Rise, Wthdoloqr, Estirrates to the Year 2100, and Research Needs. A Report of the U.S. Eh~omental Protection Zgency. Sinr>ns, Li & Associates, Fort Collins, 00. 1984. Project on Beach Sand Replenishrent. of Fedamtion. U.S. Army Corps of Engineers, 1973. Creek, Pacific Ocean to Vista, San Diego county, CA. county. Yalh, M. S., 1972. Wchan~ ‘cs of Sedkmt Transport. United States Geological Survey Water Resources Data for California, 1975-82. United States Geological my, Wo Park, CA. and Engineering, v. Van&, ed. Effect of the Santa Margarita Prepared under contract with the Bureau Flood Plain Infoxnation, Buena Vista Prepared for San Diego Pergmn Press, Oxford. .. 43 APPENDIX ONE SUBBAS INS FUTURE HYDROLOGICAL CONDITIONS CALCULATIONS . 7 com 0.0 650 6000 3500 0.05 572 0.0 HDR MDR 0.0 LDR 0.0 COMPOSIT FMST D 100% 86 86.0 CN LAG (HR) .. 0.0 86 0.3224 8 COMM 0.0 900 4000 2500 0.05 1188 HDR 0.0 MDR 0.0 LDR 0.0 COMPOSIT FMST D 100% 86 86.0 CN LAG (HR) 0.0 86 0.2117 9 COMM 0.0 450 4000 2000 0.05 594 HDR 0.0 MDR 0.0 LDR 0.0 COMPOSIT FMST D 100% 86 86.0 CN LAG (HR) 0.0 86 0.221 8 10 COMM 0.0 650 6000 2500 0.05 572 HDR 0.0 MDR 0. 0 LDR C 50% 84 42.0 COMPOSIT FMST C 50% 82 41.0 Cl% LAG (HR) . 3.0 83 0.2837 11 COMM 0.0 650 5000 2500 0.05 686 HDR 0.0 MDR 0.0 LDR 0.0 COMPOSIT FMST D 100% 86 86.0 CN LAG (HR) 0.0 86 0.2557 12 COMM 0.0 150 5000 2500 0.035 158 HDR 0.0 MDR C 40% 86 34.4 LDR C 40% 84 33.6 COMPOSIT FMST C 20% 82 16.4 CN LAG (HR) 0.0 84 0.2365 13 COMM 0.0 200 3000 1500 0.035 352 HDR 0.0 - MDR 0.0 LDR D 30X 87 26.1 COMPOSIT FMST 0.0 CN LAG (HR) PARK D 70% 82 57.4 84 0.1378 A-2 b!ATi!?SHED SUBBASIN FUTURE CHARATERISTICS 0.38 LAG = 24n( L X Lc/SQRT s) SUB- LU SOIL % BASIN 1 COMi.1 HDR F.1 D R LOR FMST 2 COMM D HDR D MDR D LDR FMST 3 COMM HDR MDR LDR FMST D 4 COMM D HDR D MDR D LDR FMST D 5 COMM D HDR D MDR LDR FMST 6 COMM D HDR D MOR D LDR FMST D 40% D 20% D 40% 40% 40% 20% 100% 30% 30% 20% 20% 50% 50% 50% 30% 20% scs C N 92 88 86 92 90 88 86 92 90 88 86 92 90 92 90 88 CN ELEV LENGTH PC-CNT BASIN SLOPE X 7, DIF L (FT) Lc (FT) n (FT/bSI) 36.8 220 3500 1800 0.05 0.0 17.6 0.0 COMPOSIT 34.4 CN LAG (HR) 0.0 89 0.2263 H- I 36.8 150 3500 1400 0.045 36.0 17.6 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 90 0.1991 0.0 700 4000 2200 0.05 0.0 0.0 0.0 COMPOSIT 86.0 CN LAG (HR) 0.0 86 0.2115 27.6 340 4700 2200 0.045 27.0 17.6 0.0 COMPOSIT 17.2 CN LAG (HR) 0.0 89 0.2394 46.0 50 3000 1500 0.03 45.0 0.0 0.0 COMPOSIT 0.0 C N 0.0 91 LAG (HR) 0.1537 46.0 150 2500 1000 0.03 27.0 17.6 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 91 0.0964 332 22 6 924 382 88 31 7 n4 . 21 COMM D 30% 92 27.6 100 10000 6000 0.03 53 HDR D 30% 90 27.0 MDR D 30% 88 26.4 LDR D 10% 87 8.7 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 90 0.4534 22 COMPI C 50% 91 45.5 80 2000 1000 0.028 211 HDR B 50% 82 41.0 MDR LDR FMST 0.0 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.0893 23 COMM D 40% 92 36.8 220 9000 4000 0.028 129 HDR D 20% 90 18.0 MDR D 30% 88 26.4 LDR D 10% 87 8.7 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 90 0.2941 24 COMM C 60% 91 54.6 85 4000 2000 0.028 112 HDR C 30% 88 26.4 MDR C 10% 86 8.6 LDR 0.0 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 90 0.1705 25 COMM C 40% 91 36.4 180 5000 2500 0.03 190 HDR C 20% 88 17.6 MDR C 30% 86 25.8 LDR C 10% 84 8.4 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 88 0.1 958 26 COMM 0.0 170 .5000 .2500 0.035 180 HDR 0.0 MDR D 100% 88 88.0 LDR 0.0 COMPOSIT FMST 0.0 CN *LAG (HR) 0.0 88 0.231 0 27 COMM C 20% 91 18.2 200 6000 3000 0.035 176 HDR C 20% 88 17.6 MDR C 60% 86 51.6 LDR 0.0 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 87 0.2663 A-4 14 COFlM 0.0 600 10000 4000 0.05 HDR 0.0 MDR 0.0 LDR C 30% 84 25.2 COMPOSIT FMST C 70% 82 57.4 C N LAG (HR) 0.0 83 0.4609 15 COHM 0.0 250 3000 1600 0.035 HDR 0.0 LDR C 40% 84 33.6 COMPOSIT YDR 0. e FtvlST 0.0 CN LAG (HR) PARK C 60% 77 46.2 80 0.1354 16 COMM B 40% 90 36.0 250 3200 1500 0.03 HDR B 40% 82 32.8 MDR 0.0 LDR B 20% 78 15.6 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 84 0.1175 17 COMM B 60% 90 54.0 250 3000 1000 0.03 HDR 0.0 U MDR B 40% 80 32.0 LDR 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 86 0.0970 FMST 18 COMM C 40% 91 36.4 250 9000 5000 0.028 HDR C 20% 88 17.6 MDR C 20% 86 17.2 LOR C 20% 84 16.8 COMPOSIT FMST 0.0 CN LAG (HR) 0.0 88 0.31 24 19 COMM 0.0 500 6000 3000 0.05 HDR 0.0 MDR 0.0 LDR C 30% 84 25.2 COMPOSIT FMST C 70% 82 57.4 CN LAG (HR) 0.0 83 0.31 96 20 COMtil C 30% 91 27.3 250 9500 5000 0.035 HDR 0.0 MDR C 30% 86 25.8 LDR C 30% 84 25.2 COMPOSIT FMST C 10% 82 8.2 CN LAG (HR) 31 7 440 41 3 440 c 147 440 139 0.0 87 0.4027 A-3 35 COMM D HDR 0 MDR D LDR FMST 36 COMM C HDR D MOR D LDR FMST 37 COMM HDR MDR D LDR FMST 38 COMM HDR C MDR D LDR FMST 39 MDR A MDR C MDR D 40 MOR C MDR D 0 41 MDR C MDR D 1 0% 20% 70% 1 0% 20% 70% 100% 40% 60% 10% 20% 70% 30% 70% 60% 40% 92 9.2 200 5000 2000 0.035 90 18.0 88 61.6 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 89 0.2057 91 9.1 200 4000 2000 0.035 90 18.0 88 61.6 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 89 0.181 1 0.0 370 4500 2200 0.035 0. 0 88 88.0 0.0 COMPOSIT 0.0 CN LAG (HR) 0. 0 0.1787 88 0.0 340 5500 2500 0.035 88 35.2 88 52.8 0.0 COMPOSIT 0. 0 CN LAG (HR) 0. 0 88 0.21 38 0.0 340 4000 2000 0.035 73 7.3 86 17.2 88 61.6 COMPOSIT 0.0 C N LAG (HR) 0.0 86 0.1638 0.0 320 3500 1500 0.035 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.1376 86 25.8 88 61.6 0.0 280 6000 3500 0.035 86 51.6 88 35.2 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.2649 A-6 21 1 264 434 326 449 483 246 28 COMM C HDR D MDR D LDR FMST 29 corm HDR MOR D LDR FMST 30 COMM D HDR D MDF? D LDR FMST 31 COMM C HDR D MDR D LDR FMST 32 COMM D HDR D MDR D LDR FMST 20% 40% 4 0% 100% 20% 30% 50% 20% 50% 30% 40% 30% 30% 91 18.2 780 4500 2500 0.035 90 36.0 88 35.2 0.0 COMPOSIT 0.0 0.0 CN 89 LAG (HR) 0.21 51 0.0 120 4000 1800 0.035 0- 0 88 88I0 0.0 COMPOSIT 0.0 CN 0.0 88 LAG (HR) 0.1918 92 18.4 150 5000 25.00 0.035 90 27.0 88 44.0 0.0 COMPOS1 T 0.0 CN LAG (HR) 0.0 89 0.2365 91 18.2 220 3400 1500 0.035 90 45.0 88 26.4 0.0 COMPOSIT 0.0 CN 0.0 90 LAG (HR) 0.1454 92 36.8 260 7000 3500 0.035 90 27-0 88 2614 0.0 COMPOSIT 0.0 CN 0.0 90 LAG (HR) 0.2933 33 COMM HDR MOR D 100% LDR FMST 34 COMFl D 50% HDR D 30% 1 DR FMST MDR D 20% 0.0 180 4000 . 2200 0.035 0.0 88 88.0 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 88 0.1916 92 46.0 180 4700 2500 0.035 90 27.0 88 17.6 0.0 COMPOSIT 0.0 0.0 CN 91 LAG (HR) 0.2205 21 1 158 158 342 196 238 202 A-5 0 49 COFlM C 40% HDR D 20% MDR 0 40% LDR FMST 50 COMM HDR MDR A 100% LDR FMST 91 36.4 280 6500 2200 0.035 227 90 18.0 88 35.2 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 90 0,2324 0.0 330 7000 3000 0.035 249 0.0 73 73.0 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 73 0.2644 A-8 L. 42 MDR C MDR D 43 MDR C MDR D 44 MDR C MDR D 45 COMM HDR MDR C LDR FMST PPRK C 46 COMM C HDR MDR C LDR FMST OPEN C 47 MDR C MOR C 48 COMM D COi4M C HDR D MDR D FCiST 60% a6 40% 88 50% 86 50% 88 409, 86 60% 88 30% 77 70% 86 10% 91 80% a8 10% 71 30% 86 702 88 40% 92 40% 91 10% 90 10% 88 0.0 240 3000 1500 0.035 422 51.6 35.2 0.0 COMPOS IT 0.0 CN LAG (HR) 0.0 a7 0.1331 0.0 220 2500 1200 0.035 465 43.0 44.0 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.1121 0.0 240 4500 2000 0.035 282 34.4 52. a 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.1871 0.0 280 4000 1500 0.035 370 0.0 23.1 0.0 COMPOSIT 0.0 CN LAG (HR) 60.2 83 0.1523 9.1 240 4500 2000 0.035 282 0.0 70.4 0.0 COMPOSIT 0.0 CN LAG (HR) 7.1 87 0.1871, 260 3900 1500 0.035 371 f. Tr V. v 25.8 61.6 0.0 COMPOSIT 0.0 CN LAG (HR) 0.0 87 0.1478 36.8 220 4500 3000 0.035 258 36.4 9.0 8.8 COMPOSIT 0.0 CN LAG (HR) 0.0 91 0.2220 A-7 ANMUAL COSTS DREDGING: LAGOOM IM EXISTING COF.IDITIOFI WATER SHED IN FUTURE COI\JDITION 191 , 500 CY SEDIMENT DELIVERY X $7.35 /CY =$1,407,525 LAGOON MODIFICATIONS: LAGOON UITH 80' LOMERIIJG WEIR, 100' HILL ST. BRIDGE & DREDGING AT 1-5 WATERSHED IN FUTURE CONDITION WEIR: OPERAT ION AND MA1 NTANENCE : HILL ST: $140,000 INITIAL COSTS LIFE: 50 YEARS = 1-5 $400,000 INITIAL COSTS LIFE: 20 YEARS = $20,000 $20,000 $2 , 800 LIFE: 2 YEARS = $1 5,000 $57,800 $30,000 INITIAL COSTS TOTAL ANNUAL COSTS: LAGOON CHANNEL DREDGED 10' DEEP 100' WIDE * WATERSHED IN FUTURE CONDITION 268,500 CY X -$15 PER CY = $4,027,500 INITIAL COSTS LIFE: 2 YEARS = $2,013,750 DETENT ION BAS INS LAGOON- IN EXISTING CONDITION UATERSHED IN FUTURE CONDITION NITH 8 DETENTION BASINS LAND: 19 ACRES X $130,000 PER ACRE = $2,470,000 LIFE: 50 YEARS = CONST'N: $1 1 , 500 EACH X MAINT: - 83 $92,000 LIFE: 10 YEARS = TOTAL ANNUAL COSTS: $49,400 $9,200 $1 30,000 $1 88,600 CREEK ENHANCEMENT LAGOON IN EXISTING CONDITION WATERSHED IN FUTURE CONDITION WITH 8 DETENTION BASINS LAND: 1,050,000 SQ FT X $4 PER SQ FT =$4,200,000 DROP STR $5,000 EACH X 160 = $800,000 LIFE: 50 YEARS = $84,000 LIFE: 5 YEARS = $1 60,000 TOTAL ANNUAL COSTS: $244,000 SIDE' CHANNEL REPAIR DROP STR $3,000 EACH X 10 = $30,000 LIFE: 5 YEARS = $6,000 TOTAL ANNUAL COSTS: $6,000 APPENDIX TWO COST-EFFECTIVENESS ANALYSIS 63 E7"MISTING BASIS 191,500 - F+"oQMB 12% 168,520 22,980 FVNRE (2a"N 26% 141,710 49,790 nrmRE CHANNEFEECT 26,810 MINIMUM F W/Dm EXISTING 20% 153,600 37,900 DGT&ENEi EXIT 45% 105,100 86,400 E"c EXIST 48,500 MINIMUM NO MlDIFICATIoN WEIR-tHILL ST+I-5 CHANNEL DGT BASINS CREm E"C E.C.R SME ARROYO GRADED AREAS AGRICUL~ S.=T SED BASIN m. SEDBASIN $1,407,525 $57,800 $2,013,750 $139,200 $160,000 $6,000 $6,000 $6,000 $11,750 $5,360 - - 22,980 $168,903 26,810 $197,054 37,900 $278,565 61,000 $448,350 1,000 $7,350 5,000 $36,750 4,500 $33,075 2,500 $18,375 1,600 $11,760 A-11 CCST- B.E" RATIO 1.00 3.0 .1 2.0 2.8 1.2 6.3 5.6 1.6 2.2 SOL'TH COAST SEDIl4EKT BASI!.! 5,000 CY SEDIMEIJT DELIVERY X $2.35 "/CY = $11,750 A!Ji.!!JAL COSTS JEFFERSO!! SEDIIXIJT BASIN 1,600 CY SEDIMENT DELIVERY X $3.35 */CY = $5,360 ANNUAL COSTS *xIrlCLl!DES REDUCTIOFI I:! PRICE OF $4/CY FOR THE VALUE OF THE MATERIAL A-1 0 ... . --a DISTANCE SECTION CROSS BETWEEN NUMBER SECTION SECTIONS AREA (SF) 100 655 101 243 102 139 103 51 9 104 526 105 645 106 520 107 581 108 384 109 1895 110 51 9 111 41 2 112 85 113 54 114 400 115 374 116 92.5 (FEET) 840 13,963 510 3,612 440 5,361 1,460 28,242 980 21,252 275 5,930 780 15,893 640 11,435 590 24,900 200 8,941 710 12,238 420 3,864 110 284, 460 3,869 640 9,170 300 2,590 171,542 TOTAL EROSION IN THE UPPER MIDDLE REACH -. .. . APPENDIX THREE MIDDLE REACH EROSION CALCULATIONS c .. _. . .I * .. 1 .. .. : .qJ.: Or ...... ; .. : ... . :- ,... .I i . 1: 1.1 ....... ;. - ... . i .... s. .. ! .. :. .. , .... .. ... .. 0' .... ... ,- .... .. 03 . , ..:a . . i.. Ii '. . ir I -? - ' .I 04 . !. ,. ..... I' ' , f:! - u ........ " 0. 0' i.... ... ... .. I ,'. ' ... ....... 4. I.. .. -. :I ,: -. .... .. 3 - ... .... .. .. - '. . ......... ..... ,.-i..; -.,... I. : *., ., I .,. I. I. .: ......... .. ti 8, ... .. ..... ..._......_ .I .. .. ... .* ... "I , . ,. ....... A-14 - ..... .- ..... - ... - - .. - ..... ....... - ai111 + , 'I A-I3 .... -. 1 /. f2 3 4 5 6 ?*St 2 I) 4 I S'78$1 2 -. 3 46 c . 'I .. i .. .... .... . I. ! .: " .. ... .. I : j i .,!.i+"i ..I.. !. : i. : .. .: ..... i:. ' .. .1 3 ; -. --.. .. .......-.... i,. . ... ! '! .. I..!., ~ ,. .... :, . _. . I .. .. ....... .. -." ,... ..... 'I . . .. ~-- . . , .-.. i . 1 .ea . .- . & # ... " .... ... ..... ., .. ... .... .... . .- . . c .1 .- I, I 1 .. ; , .: .. ... ... ! .. , .. .. ........ - . I ..... .. ... .. , ._. '. 8 -- I .! ..... 05 : 0 '. .2 .. ..... .... .! I , I. :I, .. .... .. .,- .. 11. .. .. 1 i I I.., I L/* - -- .... .1 .I : ... 1 2 I) 4 6 870aM I. 4 0 ... .. ! ....... - .. * */ I.. t . ,. . 8 .., I' : ! , I..: .. ..... ;. : .I '.. I!. " 8 ..... ,. .i ., i .q :';a ....... 3 .. ...! ... ! : I .... I. .... 0 ;..I -.,. i..:. &.-.I :... .__ :.s ... I. .. .. ... .. ........ i;;!: .. * '* - , 7""'" ._..- ~ ._ .... ...... I ,. !l;!.l, !: : : : , : 1 . i ."'. .. I. .... ;.+ '._ .. . .. /.. , .. .., ... . .. .. ... ... 'ea Q e3 ........ .............. ..... ....... ..-e-. .,. ...r*,o..,(.~ .,: .".. .Y A-15 ? 1 r: .... ... .. .. .. . - ....... - - . .......... I ........ .- ... P*zC 4. .I*Ow 1: ......... . ', - .. .. .. ..... .I 'I , : , i I I * ! e 4 I I i .. A -5 2 A-17 73 A-1 9 12-w