FIRST TECHNICAL PROGRESS REPORT DEMONSTRATION OF NON-POINT POLLUTION MANAGEMENT ON CASTRO VALLEY CREEK FOR U.S. ENVIRONMENTAL PROTECTION AGENCY WATER PLANNING DIVISION WASHINGTON DC 20460 MAY 1979 BY ALAMEDA COUNTY FLOOD CONTROL AND WATER CONSERVATION DISTRICT HAYWARD, CALIFORNIA ------- FIRST TECHNICAL PROGRESS REPORT 27940 DEMONSTRATION OF NON-POINT POLLUTION MANAGEMENT ON CASTRO VALLEY CREEK FOR U.S. ENVIRONMENTAL PROTECTION AGENCY WATER PLANNING DIVISION WASHINGTON DC 20460 MAY 1979 H. A. FLERTZHEIM, JR., DIRECTOR OF PUBLIC WORKS, ALAMEDA COUNTY PAUL E. LANFERMAN, ENGINEER-MANAGER ALAMEDA COUNTY FLOOD CONTROL AND WATER CONSERVATION DISTRICT WRITTEN BY GARY SHAWLEY-PROJECT MANAGER TECHNICAL REVIEW BY ROBERT PITT, WOODWARD/CLYDE CONSULTANTS ------- ABSTRACT This report comprises the first technical progress report for the EPA sponsored project titled "Demonstration of Non-Point Pollution Management on Castro Valley Creek," This project is part of the San Francisco Bay Area's 208 Continuing Planning Process and is the first prototype project in EPA's Region IX to be part of the Nationwide Urban Runoff Program. This report describes each of the project's work tasks and what portions of each task have been accomplished. Preliminary information based on field measurements and literature reviews is also included. As of March &, 1979 about $32,000, which is 30* of the first year's project costs ($108,000), has been incurred. As of that date the first year's work on the project is about 42% (2D weeks) complete. More than 1,3D0 person-hours have been spent on the project. The major project activities - receiving water monitoring, street surface moni- toring, and street cleaning are each about one-quarter complete and data analysis has been initiate:?. Field measurements of storrawater quality, pollutant accumulation rates, and street cleaner performance are presented in this report, but the data is not fully analyzed. ------- CONTENTS PAGE Abstract Figures Tables Section I. Introduction 1 Purpose - 1 Description of Study Area 2 Section II. Task Progress 10 Task 1 Work Plan Development 10 Task 2 Experimental Design 11 Task 3 Receiving Water Monitoring 14 Water Quality Data 21 Task 4 Street Surface Monitoring 28 Task 5 Street Cleaning and Leaf Removal Tests . 37 Task 6 Project Review Meetings 42 Task 7 Data Analysis and Report Preparation 43 Section III. Schedule 44 Section IV. Project Costs 47 Section V. Appendix 48 ------- FIGURES Number Page 1 General Location of the Castro Valley Watershed .... 3 2 Aerial View of Castro Valley 5 3 Typical Land Uses 6 4 Study Area Divisions 8 5 Runoff Flows at Monitoring Station Sites 15 6 Daily Average Discharge as a Function of Time 18 7 Precipitation & Instantaneous Discharge at Knox Station 19 8 Precipitation & Instantaneous Discharge at Seaview Sta. 20 9 Filtration at District Laboratory 25 10 Maintenance of Water Sampling Equipment 26 11 Flow Monitoring Operation 27 12 Street Surface Loading Values as a Function of Time . . 31 13 Equipment & Trailer for Street Surface Monitoring ... 32 14 Newspaper Coverage of Collection Procedure for Street Surface Samples 33 15 Compositing Street Surface Samples . 35 16 Wet Gutter & Streets 36 17 Schedule for Castro Valley Demonstration Project ... 45 ------- TABLES NUMBER PAGE 1 Rain Events During Field Activities 16 2 Summary of Monitored Receiving Water Quality Values. . 22 3 Street Surface Sample Collection Summary 28 4 Loading Values for Street Surface Samples 29 5 Preliminary Results of Street Cleaning Performance Values 37 6 Comparison of Street Cleaning Performance Values ... 39 7 Revised Street Cleaning Schedule 41 ------- SECTION I INTRODUCTION This report comprises the first technical progress report for this project titled "Demonstration of Non-Point Pollution Manage- T!ent Qr Ecitra Stl 1 e» -ree-:." "he prinzr/ acjettiva cr "tfia prt-;ect is to demonstrate the potential role of street cleaning in the manageitferit of water quality in order to help meet the 1972 Clean 'Water Act's goal cf C3ean Water. The project is being conducted by the Alameda County Flood Control and Water Conservation District (District) with 75% funding by the U.S. Environmental Protection Agency lEPA). The District's purpose is to determine if the street cleening control measure con- sidered in the local £08 program is cost-effective in improving water quality. This project is one tjf about 30 projects in EPA's Nationwide Urbav, Runoff Program. EPfc's. purpose in conducting this and the other prototype projects is to provide a means for obtaining information essential to developing a nationwide p.evactive for control Of urban runoff nrotlems. Progress on each of this project's tasks are discusswi a later section. As each tasfc progresses, data is obtained that can be used to compare the r^n'tored mass pollutant floors of tte sampled storms tfftfr the total pollutant removal of the various street clean- ing programs. This comparison will be made using a basic level of analysis. A valid data set for this analysis consists of a data point for a monitored runoff event that occurs between adjacent street surface monitoring data points- By the end of March, 1979, at least seven valid data sets have been collected. -1- ------- DESCRIPTION OF STUDY AREA Tie Castro Valley Watershed is about 5.6 sq. miles in area and is located within the San Francisco Bay Area (Figure 7). An aerial view of the watershed is presented on Figure 2. The study area is the Castro Valley Creek branch (2.4 sq. ini.) of the Castro Valley Watershed. The study area is predominantly residential. The majority of the residential land use consists of single family housing with lot sizes varying from 5,000 to 10,000 sq. ft. The estimated residential population dsrsity is approzimately 20 people per acre. Residential land use predominates at 2464 acres (70 percent), commercial Tancf use occupies about 246 acres (7 percent), and the remaining land use is'open space {B09 acres or 23 percent). Development along the stream banks within Castro Valley is intense and houses are often constructed directly over the existing stream- bed. Some light commercial areas, more than a dozen schools, and a short portion of Interstate Highway 580 are also contained within the area. Examples of typical Castro Valley lane! uses are shown on Figure 3. Topography within the drainage basin is highly variable, with land slopes ranging frora 10 percent to 70 percent in the upper end of the basin and slopes as law as 1 percent in1 the valley portion rear San Lorenzo Creek. The streambed ir the portirrs of tfs drainage basin ranges from 20 to 50 feet in width and 8 to 10 feet in depth* The streambed is often strewn with litter and debris. The study area was divided into four sub-areas (Figure 4). These horizontal divisions across the watershed, based on topography end street patterns, increase the arrffiLnt of useful data obtained from the monitoring activities. ^ ------- O 5 10 15 • -> ¦ Milei FIGURE 1. SAN FRANCISCO BAY AREA SHOWING THE GENERAL LOCATION OF THE CASTRO VALLEY WATERSHED -3- ------- Figure 2 Aerial View of Castro Valley -5- ------- -6- ------- There are many similarities in the three lower urban sub- areas. The three most important ahe the types of gutters, the shapes of the curbs and the condition of the street surfaces. Seventy-five percent of the gutters are concrete and 25 percent are asphalt. The shapes of the curbs (straight or rolled) may influence how much of the street surface contaminants are kept within the gutter and thus are available to the street cleaners, and how much is transported to the shoulder of the road and not available for pickup by normal street cleaner operation. The condition of the street surface contaminants and the performance of street cleaning equipment: 91 percent of the street surfaces in the lower three urban test areas are in fair condition, with little variability in condition or width (95 percent of the streets in these sub-areas are 20 feet to 40 feet wide). A variable that may significantly influence the quantity of nutrients that may be removed by street cleaning operations is the amount of leaf material on the streets. The largest accumulation of leaves on the streets is in the middle urban sub-area, but this dif- ference does not appear to be significant. Two important variables . that influence the effectiveness of the rain-flushing of particulates from the street surface are speed of the traffic and density of the traffic. These two variables are also very similar for the three urban sub-areas. The Castro Valley Creek branch of the Castro Valley Watershed was selected as the study area in order to reduce the study area to a more manageable size (from 5.6 sq. mi. to 2.4 sq. mi.). -7- ------- FIGURE 4. STUDY AREA DIVISIONS (CASTRO VALLEY CREEK- DRAINAGE AREA « 2.41 SQ. MI.) \ •• \ RuV &6s AREA ^fiEAVIEW AVt.STREAM CASE STATION SCALE J l"» 2000* RAIN GAGE STATION i / CASTRO VALLEY V/ATERSHEP V OUTSIDE STUDY AREA uppeiT T URBAN \ V ^M1D&4_E URBANX & feSAMPuNG AREJ! I S \ \ I I I V t RAIN GA6E STATION (C.V. FIRE 6TATIOH) 9^ «t> RAIN GAGE STATION \ WERj 0RBANJ \MPLIN(= ' A AREA f jrv# 'KNOX ST.STB GAGE STATION CASTRO VALLEY WATERSHED (CASTRO VALLEY £CMAM)T CREEKS) COM&INErP PKAINAGE AR£A»&.&9 SdJ-Ml- *255- fiAGE STATION r ------- San Lorenzo Creek, downstream of the confluence of Castro Valley and Chabot Creeks, is a large watercourse with contiguous urban development. This creek carries the flow to its discharge point into San Francisco Bay. The Castro Valley watershed is con- sidered representative of much of the residential development in the San Francisco Bay Region. -9- ------- SECTION II TASK PROGRESS TASK 1 WORK PLAN DEVELOPMENT The objective of this task was to prepare a suitable work plan describing necessary components of the project to obtain the stated objectives. This first task has been completed and it involved the preparation of the Work Plan which was submitted in January, 1979. It was based upon the original and revised submitted narrative statements. (The work plan was approved shortly after its sub- mittal.) Limited amounts of the material submitted in the work plan are included in this progress report. The following discussions describe the scope of each task, the progress towards completion (as of April 2, 1979) and preliminary results. -10- ------- TASK 2 EXPERIMENTAL DESIGN The objective of this task was to determine an appropriate balance between sampling effort and expected resultant precisions based'on the available project funds. This task was completed during the development of the work plan. It involved determining the optimal sampling and testing pro- cedures to be used in the evaluation tests. This discussion describes the procedures used to calculate the necessary number of street surface subsamples and the number of storms needed. The number of storms monitored is most important. Final data analysis is dependent on monitoring representative storms having large variabilities in rainfall characteristics. An estimated twenty monitored storms are necessary to obtain an 80% confidence level that monitored storm yields would be different for a monthly versus a daily street cleaning program. The analytical procedure used to determine the number of street subsamples needed involved weighing individual subsamples in the study area to calculate the standard deviations (cr) and the means (x) of the street surface loading values. From these two values, the number of subsamples necessary (N), depending on the allowable error (L), was determined. An allowable error value of about 25 percent, or less, was used. The formula used (after Cochran 1963) is: N = 4o-2/L2. - With a 95 percent confidence limit, it determines the number of samples necessary to determine the true value for the loading with a range of +L. -11- ------- Individual samples were taken at 100 locations in the three study areas to determine the loading variability. The loadings were found to vary within the study area but the median values in the three test areas, were fairly close. The overall minimum loading measured was about 50 Ib/curb-mile, the overall maximum value was about 3000 lb/curb-mile, and the overall median value was about 400 Ib/curb-mile. The median values in the three areas were about 320, 540 and 470 lb/curb-mile. The following table summarizes the number of sub-samples necessary for the three test areas and for several allowable error values: f .NUMBER OF SAMPLES REQUIRED IF THE ALLOWABLE ERROR IS: Study Area 5% 10% 25% 50% 100% Lower-urban 400+ 100 20 5 3 Middle-urban 400+ 250 36 8 4 Upper-urban 400+ 200 25 _6 _3 Total urban area 1200+ 550 81 19 10 The most rigorous sampling program that could be conducted using a single sampling team was therefore chosen based on an allow- able error of 25 percent. This allowable error prercentage was chosen to keep the precision and the sampling effort at reasonable levels. The three test areas can be sampled in less than 5 hours. The data were also examined to determine if the study areas should be divided into meaningful test groups. As described in the study -12- ------- area description section of this work plan, the only major differences between the test areas was topography. The test areas were therefore generally divided on the basis of topography. The total amount of street surface particulates removed during each test is insignificant when compared to the total street surface loadings in the whole test area. (Generally, the sample would be 0.1 percentof the total street surface loadings for the area.) -13- ------- TASK 3 RECEIVING WATER MONITORING The objective of this task is to monitor washoff quantities in Castro Valley Creek during a variety of rainstorms during the project period. Runoff event monitoring commenced on December 5, 1978, with the completion of installation by USGS of two receiving water monitor ing stations located at Seaview Avenue ('USGS #11181004) and Knox Street (USGS #11181096). Figure 5 shows some of the monitored flows at these two locations. To meet USGS fiscal year budgeting require- ments, the District will be contracting with USGS in May for the following year's laboratory services. To date, 15 individual storms have been monitored. However for the majority of the data analysis purposes, a monitored storm cannot be used unless a street surface sample is collected from the entire study area before and after that specific storm. If an unmonitored storm occurs in a series of monitored events between adjacent street surface tasks, the complete runoff yield for that storm series cannot be calculated to be compared to the differences in street loading and the initial street loading before the runoff event. So far,seven valid data sets have been obtained for use in the basic level of analysis this first project year. Since the onset of the 1978/79 wet season measurable rain has occurred on 39 days (with 4 days of rain prior to initiation of field activities). These events are summarized on Table 1. The total amount of rain from September, 1978 to April 17, 1979, has ------- FIGURE 5 RUNOFF FLOWS AT MONITORING STATION SITES A. Discharge at Knox Street Station B. Discharge at Seaview Avenue Station C. Discharge and Staff Gage at Seaview Avenue Station -15- ------- TABLE 1 RAIN EVENTS DURING FIELD ACTIVITIES-^ Date Total Duration Average Intensity Peak Intensity (inches) (hours) (inches/hour) (inches/hour) Dec. 17, 1978 * .39 12.5 .03 .14 Dec. 18 .05 15.75 .003 .03 Dec. 19 .01 0.25 .01 .01 Jan. 3, 1979 .10 3.25 .03 .03 Jan. 4 .03 9.25 .003 .02 Jan. 5 .01 0.25 .01 .01 Jan. 7 * .34 14.75 .02 .05 Jan. 8 * 1.24 6.0 .21 .40 Jan. 9 .18 8.25 .02 .04 Jan. 10 * .78 4.25 .18 .39 Jan. 11 1.80 20.75 .09 .27 Jan. 14 * 1.43 20.75 .07 .33 Jan. 15 .28 12.75 .02 .09 Jan. 17 .24 5.75 .04 .11 Jan. 30 .01 .25 .01 .01 Feb. 3 .01 .25 .01 .01 Feb. 13* 1.11 13.25 .08 .25 Feb. 14 .09 9.25 .01 .01 Feb. 15 .01 .25 .01 .01 Feb. 16* .49 11.75 .04 .21 Feb. 17 .01 .25 .01 .01 Feb. 18 .45 8.75 .05 .20 Feb. 19* .06 1.25 .05 .05 Feb. 20* .74 17.75 .04 .25 Feb. 21 * .41 11.0 .04 .17 Feb. 22* .64 13.25 .05 .30 Feb. 23 .18 23.25 .01 .07 Feb. 25 .07 2.75 .03 .04 Feb. 26 .09 9.50 .01 .06 Feb. 28* .58 6.25 .09 .17 1/ Proctor School Rain Gage, USGS # 71-1810.08 * Monitored Events -16- ------- been 19.4 inches, as compared with an annual average of 21.7 inches. These individual rains have lasted from 15 minutes to 23 hours. Yet, for analytical purposes, the storm periods have ranged to a maximum of 7 days. The storms of most concern (relating street cleaning activity to stormwater quality) are expected to be within the range of 0.1 to 0.5 of an inch. Storms larger than this may have large erosion yields. The following is a discussion of the rain characteristics of the storm periods which constitute the first five valid data sets. Within these valid data sets the minimum amount of rain has been 0.39 of an inch and the maximum has been 2.8 inches. Peak intensities within these individual storm periods range from a low of 0.14 in./hr. to a high of 0.4 in./hr. Average intensities have ranged from 0.04 to 0.23 in./hr. Data analysis may be more difficult with some of the storm data due to masking effects of erosion (erosion may deposit soil onto the streets and hinder the comparison of runoff yield with before and after street loadings). Figure 6 is a USGS computer plot of average daily water discharge as a function of time measured at the Knox Street station. The ordinate is a logarithmic discharge scale and the abscissa is time. The maximum daily averaged flow shown is 37 cfs. Figures 7 and 8 are USGS plots of precipitation and instantaneous discharge at Knox and Seaview stations respectively of February 16-22. -17- ------- FIGURE 6 DAILY AVERAGE DISCHARGE AS A FUNCTION OF TIME CASTRO VALLEY CREEK AT KNOX STREET • — ¦ — -1 ¦ *— - — .. . I 1 — j 1 ~r I t 1 1 i I — -J 1 1 __j I .. u_. 1 ¦ 44- 1 i ¦ - -!—| • -4— r ;4~ . J 1-4 — hH —f— 1 — t 1 —t 1 " L_ — .X. _iL_~ .. 1 I It —I —1 — —i— i ^r- T Z - i !| m ¦%— I 1 ! - 1 ¦ i i 1 r- ¦- — • — 1— 4 - — — 4— - i— ' 1 w — - - ~i • J_ _ --J-—r -1 J . . ». 1 — I i T I ¦ r 7 i i I i i U .1 j—» —1 1 i . 1 — 1 4- ' :: i J. —J-4 — 1— I— r- .. 1. " r-i c -- H i 1 1 1 1 » — i 1 1 1 i % T — -i 1 i 1 k • 1 t 1 t I au • » i. I 1 lb 11 » a » K » • » * L * it ii » h h I ii k a » » i ¦ Oct Nov Dec Jan Feb Mar -18- ------- FIGURE 7 PRECIPITATION & INSTANTANEOUS DISCHARGE AT KNOX STREET 0=PRECTPITBTION AT SYDNEY SCHOOL GAG£ X=PR£CTPITRTION AT PROCTOR SCHOOL GRGE INSTRNTRNIEOUS DISCHARGE 11181006 CflSTRO VRLLEY C FIT kNOX ST ------- FIGURE 8 PRECIPITATION & INSTANTANEOUS DISCHARGE AT SEAVIEW STREET 0=PRECIPITRTION RT SYDNEY SCHOOL GAGE X=PRECIPITflTION RT PROCTOR SCHOOL GRGE INSTRNTRNEOUS DISCHARGE 11181004 CRSTRO VALLEY C RT SERVIEW RVE ------- Water Quality Data Table 2 presents the water quality values measued from initiation of project field activities through January 14, 1979, (last date of sample analysis received from the laboratory). The values shown are for hourly composite samples. The values not yet received from the lab will be for total storm composites. The reason for taking composite instead of discrete samples is to save considerable laboratory and labor costs while still determin- ing mass emissions permonitored event. The samples at both the Seaview Avenue station (rural) and the Knox Street station (urban) were taken by ISCO automatic samplers (Model #1680) and are controlled by ISCO bubble-type flow meters. The samplers are set to take samples at predetermined flow increments. USGS personnel checked the results from the laboratory as part of the quality control process of the cooperative agreement betv/een the District and the USGS. Values for lead are not reported here but will be made available at a later date. The USGS central laboratory in Denver had a problem with contamination of their acid preservatives and it was decided to verify the accuracy of the data before it was released. Probably the most important progress of the project has been the education and experience acquired by the District's per- sonnel in performing water resource investigations. For example, the District personnel have never worked with the ISCO automatic sampling equipment before and now after one season of monitoring, they are experienced at correcting mechanical problems which occur -21- ------- TABLE 2 SUMMARY OF RECEIVING WATER QUALITY VALUES MONITORED: November 1978 - January 14, 1979 Knox and Seaview Stations on Castro Valley Creek Total Solids S04 SO. MONITORING Total Non 01s. Ois. STATION Temp - Sp. Solids Pb Zn As SOj tOD TKH HH. TP Filter Cations Anions Turbidity AND DATE cC Cond. pH (mq/1) (u9/l) (ug/1) (ug/1) (mg/f) (ng/1) (mg/1) (mg/1) (hh/1 ) (mg/1) (MEQ/1) (HEQ/T) (NTU) Knox 12-17-78 0645 10.5 111 7.2 395 250 1 52 200 3.5 .03 .74 130 0 1.083 31 0745 125 6.8 0845 10.5 320 7.0 143 90 2 21 83 1.2 .01 .38 34 0 .437 20 0945 10.5 459 7.4 273 100 2 49 83 1.1 .02 .46 18 0 1.020 14 1045 11.0 233 7.4 321 100 2 62 1145 11.0 233 7.2 261 140 2 37 110 1.4 .02 .46 69 0 .770 26 1245 11.0 96 7.1 288 220 3 27 130 1.8 .02 .48 134 0 .562 37 1345 11.0 151 7.5 240 170 3 13 110 1.4 0.05 0.40 150 0 .271 25 1445 11.0 158 7.2 222 100 4 21 59 1.5 0.02 0.41 113 0 .437 26 1545 11.0 110 7.0 200 130 3 15 93 1.5 0.03 0.41 121 0 .312 32 Knox 1-7-79 1800 11.0 133 7.7 125 70 54 .97 .03 .22 42 29 1900 11.0 127 7.7 126 80 48 .99 .04 .25 31 30 2000 11.0 108 7.6 111 110 44 .73 .05 .21 33 33 2100 11.0 136 7.5 116 30 32 .67 .03 .23 35 16 2200 11.0 178 7.5 138 100 39 .66 .01 .21 11 10 2300 11.0 202 7.6 131 100 43 .81 .03 .23 30 13 Knox 1-8-79 1030 11.0 96 7.2 361 430 140 3.2 .07 .58 246 84 1130 10.0 70 7.5 412 a) 340 110 2.1 .07 .66 312 108 1230 11.0 53 7.4 856 a "C 520 140 3.3 .09 .92 552 156 1330 11.0 108 7.2 710 &. 300 120 3.4 .11 .88 572 140 1430 10.0 169 7.5 342 CI +> 180 26 1.9 .12 .61 ISO 68 Seaview w 1-8-79 « 1230 10.0 153 7.6 1990 580 220 7.6 .39 2.1 1420 400 1330 10.0 273 7.4 766 m 200 100 3.4 .22 1.0 524 320 1430 11.0 342 7.3 497 ¦s 170 74 2.4 .13 .88 298 136 Seaview t 1-10/1-1V -79 R 2200 11.0 156 7.6 1340 t 250 608 2300 11.0 148 7.6 4300 ft 90 624 2400 11.0 210 7.6 292 70 432 0100 11.0 233 7.6 1640 O 4-> 60 352 0200 11.0 168 7.5 684 trt 130 312 0300 11.0 343 7.5 632 0) » 110 280 0400 11.0 571 7.5 584 * 100 264 0500 11.0 479 7.7 500 90 203 0600 12.0 323 7.7 934 m 180 264 Knox0700 12.0 418 7.7 954 160 192 1-10/1-11 -79 2110 12 84 8.0 380 180 40 2210 10.5 114 7.6 327 220 «4 2310 12 126 7.5 311 120 64 0010 12 172 7.6 296 120 39 0110 11 82 7.7 200 110 36 0210 12 146 7.7 278 110 52 0310 11 220 7.7 256 110 54 0410 11 271 7.6 274 110 58 0510 12 226 7.5 262 110 40 0610 12 223 7.5 336 120 74 0710 13 326 7.9 343 110 68 Seaview 1-14-79 0500 9.0 263 7.3 364 50 72 0600 9.0 316 7.2 369 60 116 0700 9.0 287 7.2 381 70 120 0800 9.0 1073 7.2 518 120 272 0900 9.0 439 7.3 643 140 240 Knox ¦ 1-14-79 0400 10.0 313 7.3 201 ISO 48 0500 9.0 91 7.0 175 110 33 0600 9.0 442 7.2 152 100 36 0700 9.0 145 7.2 169 90 44 0800 9.0 553 7.2 175 80 4? 0900 9.5 247 7.2 178 100 32 1000 9.5 346 7.4 182 80 37 -22- ------- in the equipment. An example of the level of skill acquired by District personnel is that USGS has recently asked the District to provide training in the operation of the ISCO automatic equipment to an adjoining county. This acquisition of water resource investi- gation skills will benefit this project during the second year in terms of improved efficiency. The use of these newly acquired skills is illustrated by figures on the following pages. These figures, photographed at the District's basement laboratory, show preparation of water samples for shipment to the USGS laboratory for chemical analysis and the maintenance of the water sampling equipment. -23- ------- FIGJRE 9 FILTRATION AT DISTRICT LABORATORY Principle Investigator and Flood Control District Personnel Filtering Runoff Samples for Analysis Engineer/Scientists Prepare Water Samples for Shipment to USGS Lab in Denver -25- ------- FIGURE 10 MAINTENANCE OF WATER SAMPLING EQUIPMENT Removing Top of Automatic Sampler at Knox Street Station Capping Water Sample Bottles in Preparation for Transporting to Lab -26- ------- FIGURE 11 FLOW MONITORING OPERATION Checking Flow Recorder for Time of Sample Collection ****** Flow Monitoring Equipment, left to right, Manometer, Flow Meter, and Flow Recorder -27- ------- TASK 4 STREET SURFACE MONITORING The objectives of this task are to monitor street surface particulate loadings to supply information for comparison with the runoff yields and to monitor the effectiveness of the street clean- ing programs. Table 3 summarizes the number of street surface samples collected between December 4, 1978 and March 7, 1979. All of these samples have been analyzed by particle size for chemical constitu- ents. Table 4 lists the loading values of the street surface contaminant samples. f- TABLE 3 SAMPLE COLLECTION SUMMARY Number of Type of Sample Samples Collected Leaf Removal 12 Street Cleaning 16 Accumulation 34 L. TOTAL 62 Loading of Contaminants The preferred approach to determining loading values of street surface contaminants for a given area is by direct sampling because this method considers the site-specific conditions of a given area. Previous research studies (URS, 1974) have found that geographical location (climate category), land use, street surface condition and type of adjacent landscaping are the most significant -28- ------- TABLE .4 loading values for street surface samples imple Date ~Sample ID# Lbs/Curb Mile Date ID# Lbs/Curb 12/ 4/78 L- 1L 936 1/25 :S- 3M (before) 703 12/ 4 L-2M 524 1/25 S- 4M (after) 473 12/ 4 L- 3U 526 1/25 A-16L 927 12/ 5 L- 4L 437 1/25 A-17U 466 12/ 5 L- 5M 586 21 1 A-19L 465 12/ 5 L- 6U 524 2/ 1 A-21U 425 12/ 6 L- 7L 527 2/ 1 S-20M (before) 462 12/ 6 L- 8M 878 2/ 1 S- 6M (after) 455 12/ 6 L- 9U 475 2/ 7 S-23M (before) 673 12/ 7 L-10L 789 2/ 7 S- 8M (after) 429 12/ 7 L-11M 724 2/ 7 A-22L 703 12/ 7 L-12U 541 2/ 8 A-24U 577 2/15 A-25L 444 2/15 A-26M 360 2/15 A/27U 703 2/23 A-28L 637 12/15 A- 1L 769 2/23 A-29M 335 12/15 A- 2M 1108 2/23 A-30U 258 12/15 A- 3U 750 2/27 A-32M 567 12/20 A- 4L 700 2/27 S-33U (before) 488 12/20 A- 5M 843 2/27 S-12U (after) 198 12/20 A- 6U 546 2/28 A-31L 903 12/28 A- 7L 978 2/28 S-10U (before) 332 12/28 A- 8M 1069 2/28 S-11U (after) 202 12/28 A- 9U 841 3/ 2 A-34L 1646 1/10/79 A-10L 420 3/ 2 S-35M 636 1/10 A-11M 335 3/ 2 S-36U (before) 236 1/10 A-12U 485 3/ 2 S-14U (after) 249 1/16 A-13L 470 3/ 5 S-15U (before) 298 1/16 A-14M 494 3/ 5 S-16U (after) 193 1/16 A-15U 427 3/ 7 A-37L 896 1/19 S- 1M (before) 454 3/ 7 A-38M 657 1/19 S- 2M (after) 381 3/ 7 A-39U 271 L = leaf removal test samples A = accumulation test samples S = street cleaning test samples ~Location U = upper area M = middle area L = lower area -29- ------- factors making up loading values. The data concerning traffic effects were not clear; some studies, notably Shaheen (1975) found a correlation between pollutant loadings and automobile traffic for individual cities. Loadings Over Time Figure 12 shows the street surface loading values as a function of time. In general, the pattern of data illustrates the sawtooth pattern associated with the deposition and removal of particulates. The highest loading measured was 1646 lbs/curb mile in the lower subarea and the lowest loading was 193 lbs/curb mile in the upper area. This minimum value represents the cleanest that any of the subareas have been after being cleaned either by rain or by street cleaning. It is interesting to note that this loading was a result of street cleaning and was cleaner than the rains had gotten the area. For the most part, the upper subarea was the cleaner of the three subareas. The maximum loading values were measured in the lower subarea. This probably results from the fact that as of March 7, no street cleaning had been conducted in the lower subarea. -30- ------- 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0 Fig, 12 - STREET SURFACE LOADING VALUES AS A FUNCTION OF TIME Legend UPPER SUBAREA Time, Days ------- Figure 13 shows the equipment used to monitor the street surface contaminants. This equipment consists of two 2.5 HP vacuums with accessories, one 5000 watt generator and an equipment trailer. Figure 14 is one example of local media coverage of the project. Publicity such as this has been found to be a good way to communicate project activities to study area residents. Other progress in the publicity area has included a newspaper article and a television interview. Figure 15 shows the compositing of all the street surface samples in preparation for shipment to the lab for chemical analysis. FIGURE 13 EQUIPMENT AND TRAILER FOR STREET SURFACE MONITORING -32- ------- SECOND SECTION Friday, February », W79 13 FIGURE 14 Newspaper Coverage of Collection Procedure for Street Surface Samples Tidying up the town The process looks like a' 'Keep Ameri- ca Beautiful" commercial gone berserk — crews of men driving big street sweepers down Castro Valley streets, and then carefully vacuuming up the bits of dirt they missed. But it's really a project for cleaning up the bay. Alameda County Flood Control officials are trying to find out if sweep- ing the streets will keep oil, dirt and other gunk from washing into local streams, and eventually into the bay, during rainstorms. Fred Wolin, an engineer-scientist with the depart- ment, vacuumed a section of Knox Street during this week's evaluation of the program by federal Environmen- tal Protection Agency officials. -33- ------- FIGURE 15 COMPOSITING STREET SURFACE SAMPLES Consultants Compositing Street Surface Samples for Shipment to Laboratory -35- ------- FIGURE 16 WET GUTTER AND STREET Runoff Ponding in Rolled Gutter Near Seaview Avenue Station The above figure illustrates a street surface sample problem. Extensive gutter flow and ponding on the street can occur for long periods of time after rain has ended. For analytical purposes, street surface samples should be obtained at the earliest possible time after a storm has been monitored. A stormwater sample cannot be used in the complete data analysis unless a street surface sample is obtained from the whole study area before and after that specific rain. Street sampling locations need to be periodically moved to avoid these condi- tions. If the sampling station cannot be conveniently moved, then the sample strip must be shortened before the gutter is reached. Very little particulate pollutants would be in the curb if the curb was covered with flowing water for long periods of time. -36- ------- TASK 5 STREET CLEANING AND LEAF REMOVAL TESTS The objective of this task is to measure the effectiveness of various standard street cleaning programs and leaf removal practices. As of March 2, 1979, eight street sweeping tests and twelve leaf removal tests have been conducted. The results of these street cleaning tests are shown on Table 5. Of the two removal measures shown, the preferred one is the pounds per curb mile removed measure (unit removal rate) rather than the percent of the before loading removed. The pounds per curb mile removal value is necessary in designing a program to meet a goal of removing a certain number of pounds of pollutant in a given area. A pounds per curb mile removal value is necessary for this study in order to perform mass balance calculations and to compare pollutant removals from the study area by street cleaning and storm events. The preliminary results so far indicate relatively good pollutant removals. The three lowest removal values shown on Table 5 (73 lbs., 7 lbs. and negative 14 lbs. per curb mile) were probably TABLE 5 PRELIMINARY RESULTS OF STREET CLEANING PERFORMANCE VALUES Location Date of Test Amounts Removed Percentage of Before (lbs/curb mile) Loading Removed (%) Middle Subarea 1/19/79 73 16 Middle Subarea 1/25 231 33 Middle Subarea 2/ 1 7 1 Middle Subarea 21 7 245 36 Upper Subarea 2/27 290 59 Upper Subarea 2/28 131 39 Upper Subarea 3/ 2 -14 -6 Upper Subarea k. 3/ 5 105 35 -37- ------- caused by the street cleaning equipment not being maintained at optimum condition. On the January 19 test, the left gutter broom required maintenance and similarly on February 1, the main pickup broom was worn to the maximum extent. On the next test, 6 days later, an improvement to over 200 lbs/curb mile pickup occurred because of replacement of this broom. The lowest removal value occurred on the March 2 test, when the operator noted 15% wear on the main pickup broom and when the street surface loadings before the test were very low. Table 6 compares street cleaning performance data for total solids loading from this demonstration project with previous values obtained from the development of the County's 208 Plan and from the completed San Jose street cleaning demonstration project. As can be seen, this year the amounts removed from Castro Valley were higher than the previous Castro Valley and Oakland data. The removal value obtained in the Oakland industrial area was higher, but the initial loading was much higher. The removal values for San Jose's good asphalt areas were lower. The poor condition asphalt streets and oil and screened surface streets in San Jose had greater removal values (due to greater initial street loading values). -38- ------- TABLE 6 COMPARISON OF STREET CLEANING PERFORMANCE VALUES Date Amount Removed (lbs/curb mile) % of Before Loading Removed Castro Valley-Nil/average Castro Valley-U-average Castro Valley-R-^average 1979 1979 1977 140 130 23 22 32 Oakland-R^/average Oakland-ii/one test only 1977 475/77 50 380 5 14 San Jose--^ asphalt streets-average 1976-1977 San Jose-poor asphalt streets-average 1976-1977 San Jose-oil and screened surfaces streets- average 1976-1977 104 540 170 37 40 ]_/ Middle Study Area; upper study area 2/ Residential Area, Source: Alameda County 208 Plan Zj Industrial Area, Source: Alameda County 208 Plan 4/ Industrial Area, Source: Alameda County 208 Plan 5/ San Jose Demonstration Project, Source: Pitt -39- ------- Street Cleaning Test Schedule Table 7 shows a revision of the preliminary street clean- ing schedule. This schedule was revised due to the cancellation of the regenerative air type of street cleaning equipment (Tymco) tests. The Tymco equipment was cancelled due to financial consid- erations. Essentially the.revision moved up the second and third phases of Mobil cleaning so as to obtain a complete set of data for the Mobil street cleaner operating in all three study areas. -40- ------- 3/8/78 Shawley TABLE 7 REVISED STREET CLEANING SCHEDULE^^ 5-DAY Work-Weeks Upper Urban Area Middle Urban Area Lower Urban Area 11/20 11/24/78 11/27 12/1 12/4 12/8 12/11 12/15 12/18 12/22 12/25 12/29 1/1 1/5/79 1/8 1/12 ,(2) 4L 0 0 0 0 0 0 0 0 0 0 0 0 fs) 4L* 0 0 0 0 0 0 1/15 1/19 1/22 1/26 1/29 2/2 2/5 2/9 2/12 2/16 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 2/19 2/23 2/26 3/2 3/5 3/9 3/12 3/16 0 3 0 0 0 0 0 5 0 > 0 0 0 3/19 3/23 3/26 3/30 4/2 4/6 4/9 4/13 4/16 4/20 4/23 4/27 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 5 4/30 5/4 5/7 5/11 5/14 5/18 5/21 5/25 5/28 6/1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 (1) All performed w/Mobil - number of street cleaning tests per week are shown. (2) Not monitored; a starting date (3) Leaf removal tests -41- ------- TASK 6 PROJECT REVIEW MEETINGS The objective of this task is to communicate preliminary findings and to obtain guidance from various interested parties. The first meeting with the EPA project officer was held in January 1979 in San Francisco to discuss the work plan. The next meeting is scheduled in May to review this technical progress report. A meeting, late in February, was held with the project Technical Advisory Committee members. These members represent the Regional Water Quality Control Board, the Corps of Engineers, the U.S. Geological Survey, ABAG, and the District. An EPA representa- tive was also present. The purpose of the meeting was to review the project work plan. The discussion centered on the street cleaning regime and the consequent division of the study area into three subareas. The major reason for this division was that the study area, even at two square miles, was too big to sample and clean in one day. Without the division there would not have been sufficient time for sampling the streets on the day of street cleaning, and would have exceeded available project resources. Another reason is that the division will allow us to study the effects of the different topography and different land uses of the subareas. The result of the meeting was the decision that during the second project year a more direct scheme of cleaning the entire study area at one time would be used. The Technical Advisory Committee will probably next meet in July (after pre- liminary Jata analysis is underway) to review this progress report and resultant data available at that time. -42- ------- JASK 7 DATA ANALYSIS AND REPORT PREPARATION The objective of this task is to analyze the data in accordance with the objectives of the study and to present the results in a concise and readable manner. The data analysis has just started and some preliminary data and is shown in tables and figures of this report. The runoff and street surface data collection effort will continue during the dry season; however, this dry season data will not be reflected in the data analysis of the Annual Report. The first formal report (i.e., the Work Plan) was completed in January 1979. This report is the first technical progress report and discusses the project progress to April 1979. The next report will be the Annual Report available for review in September of this year. This Annual Report will cover the task progress and detailed data analysis from the initiation of the project in November 1979 through data collected to June 1979. This Annual Report will also constitute the Work Plan for second year activities. The Annual Report is expected to be published by EPA. The next progress report will be submitted in January, 1980 and the final report draft will be submitted, in September, 1980. -43- ------- SECTION III SCHEDULE The proposed schedule for conducting the tasks of the demonstration project is shown on Figure 16. The Work Plan was approved in early February, 1979, and the experimental design (Task 2) was completed as part of the work plan development. The two receiving water monitoring stations (of Task 3) became operational in December, 1978. This runoff monitoring will be kept operational throughout the dry season until October to allow for monitoring of possible summer storms. Base flow monitoring will be performed in May to compare base flow quality with stormwater quality. Task 4, street surface monitoring will also continue until October for the first year of the project. Between June and September, samples will be taken every other week. In September, weekly samples will be taken. Task 5, street cleaning and leaf removal tests began in early December with twelve leaf removal tests conducted. Full-scale street cleaning tests began on January 19, 1979. These full-scale tests will continue until June using the street cleaning schedule presented in this report. The use of a regenerative air street cleaner was cancelled in March due to budget constraints. -44- ------- FIGURE 1 7. SCHEDULE FOR CASTRO VALLEY DEMONSTRATION PROJECT t A s k s SEP OCT NOV I DEC 1979 fes I mn urn wr juh | »UG OCT DEC 1980 ft* «n |Mr JUL SEP OCI EntRinorrAL ksig* ww pu» Dcmwmi receiving hah* mutton iig srarr surface rwrrofltNG S1BBET tLENIItt t LEAF ROWM. ItSTS miECT Knot rcnwss BATH MMLYSIS i REPORT PKPARATIOH -l I J 1 ' ¦ ' J I 1 I » * ' la/n/n incura tnnur itnan UH M MFC i wa *«ammcr «i rn uigs m nu m tccuviim wtTcn mimira to >e cncimo to ntir wct r.r. •udgetim COOTMCT (W TtM TM ftEOUtKB ------- The first EPA project review meeting in February consisted of work plan review and approval. The second formal meeting is scheduled for mid-May to discuss this progress report and the possi- bility of additional project funding. The third review meeting is scheduled for the first of September to discuss the Annual Report and continuation of second year efforts. Tentatively, the second year's first progress report would be submitted in January 1980 and the final report would be available in September, 1980. It is imperative to have signed contract agreements with ABA6/EPA no later than September 1 to keep work on schedule without any lapse of time between first and second year activities. -46- ------- SECTION IV PROJECT COSTS The first year's project costs were estimated to be $108,000; $75,000 was to be funded by EPA and $33,000 matched by the Alameda County Flood Control and Water Conservation District. As of March 8, 1979, (20 weeks and 42% of the way through the project), the total estimated project expenditures were about $32,145 (30% of the estimate). EPA costs have totaled $20,202 (27% of $75,000). The Flood Control District's match totaled $11,942 (36% of $33,000). ABAG, the project sponsor, has been credited its $5,000 share of the first year's project cost with services yet to be rendered. More than 1,300 person-hours have been spent by District personnel and consultants on the project as of March 8, 1979. The following list summarizes actual expenditures and budget estimates to March 1 , 1979, by task: Task Title Actual/Budget % 1 Work Plan Development $2,825/ 3,500 = 81% 2 Experimental Design $2,848/ 1,500 = 190% 3 Receiving Water Monitoring $8,914/38,000 = 23% 4 Street Surface Monitoring $7,480/26,060 = 29% 5 Street Cleaning and Leaf Removal Tests $2,229/12,000 = 19% 6 Project Review Meetings $1,234/ 1,900 = 65% 7 Data Analysis and Report Preparation $1,200/20,000 = 6% -47- ------- SECTION V APPENDIX The raw data is riot included here. It is too lengthy to reproduce in this interim report. It is available for viewing at the District offices in Hayward. The raw data will be included in the September Annual Report. -48- ------- |