 |
|
Effectiveness of
Street Sweeping for Floatables Control INTRODUCTION The New York City (NYC) Department of Environmental Protection (DEP) initiated a CityWide Floatables Study in June 1989 as a comprehensive investigation of floatable pollution. Information developed during that study (HydroQual, 1993) indicated that much of the material found on area shorelines resembles items discharged by combined sewer and storm sewer systems. Furthermore, materials exiting sewer systems have characteristics similar to street litter. Based on those results, alternatives to control the quantity of litter and floatables on New York City (City) streets would also help control the quantity of floatables discharged into New York Harbor. A second floatables study, begun in October 1991, evaluated three basic approaches for the control of floatables: street cleaning for reduction of floatables accumulation on streets, catch basin design and operation to minimize releases to the sewer system, and containment booms placed in waterways at strategic locations as an endofpipe measure to protect beaches and other sensitive areas. The purpose of this paper is to present results from the second floatables study (HydroQual, 1995) which pertain to the effectiveness of manual street sweeping for control of floatables. Specifically, this paper focuses on the results of a pilot program conducted by the City during the summers of 1993 and 1994. In addition, this paper examines the effects of almost continuous sweeping provided by private contractors in other areas of the city. NEW YORK CITY STREET CLEANING PRACTICES The New York City Department of Sanitation (NYCDOS) has the responsibility of cleaning City streets. City streets are cleaned for aesthetic, health and quality of life reasons. Although sweeping street litter and transporting it to landfills prevents that litter from entering the storm and combined sewer systems, and hence precludes it from being discharged to waterways, floatables control has not been considered an explicit goal of street sweeping. Currently, there are approximately 500 NYCDOS employees involved with the task of cleaning City streets. These cleaners operate mechanical sweeper trucks to remove street litter from nearly 51% of the 18,800 curb kilometers in New York City. Mechanical street sweeping frequencies vary throughout the City; some areas are mechanically swept as many as three times each week. Over the last 25 years, there have been up to 2,200 employees involved in street cleaning activities, many of whom were deployed to perform manual sweeping. Except for special cleaning operations, such as after parades and special events, the NYCDOS has discontinued manual sweeping practices. For over a decade, private selftaxing organizations known as Business Improvement Districts (BIDs) have been operating to upgrade their neighborhoods and to augment City security and sanitation services. Currently, approximately 36 BIDs operating in the City of New York conduct some level of manual sweeping on streets and sidewalks. Sanitation information available for 24 of the BIDs indicates that BID sizes vary from about 800 curb-meters to over 22,800 curb-meters and together account for about 105 kilometers or 0.5% of the City's total curb length. During warm weather, many of these BIDs conduct manual sweeping anywhere from 3.5 to 16 hours each day, five to seven days per week. ENHANCED CLEANING PILOT STUDY In order to evaluate the ability of manual street sweeping to reduce street litter and floatables, the New York City Department of Environmental Protection funded the hiring of temporary sanitation street cleaners for a four month pilot study during the summer of 1993. Under the direct supervision of the Department of Sanitation, these street cleaners were deployed over a 180 hectare square kilometer area in the Brooklyn community of East New York. This densely populated area of historically high litter levels is served by combined sewers and represents the entire drainage basin of Hendrix Creek, a tributary of Jamaica Bay (Figure 1).
Normal, or "Baseline," NYCDOS street cleaning
practices in the pilot study area consist of two mechanical sweeps a week for each
blockface. The pilot program introduced two levels of enhanced street cleaning, each of
which augmented the Baseline cleaning with manual sweeping of the streets and of the
"broomline" portion of the sidewalks (that is, the portion of the sidewalk that
the cleaner can reach from the curb with a broom). "Level 1" cleaning augmented
the Baseline cleaning with manual sweeping so that each blockface was swept a total of
once per day, six days a week. "Level 2" cleaning added an additional daily
manual sweeping to double the cleaning frequency to two sweeps per blockface each day, six
days a week. To minimize seasonal variations, Level 1 and Level 2 cleaning were staggered
monthly during the summer of 1993, and Baseline cleaning was monitored during the summer
of 1994. Test periods and mechanical and manual sweeping frequencies are summarized in
Table 1 for the Baseline, Level 1 and Level 2 cleaning scenarios.
In order to quantify litter amounts present under each of the three cleaning conditions described above, a field monitoring program was conducted. Once each week, all litter was collected from each of 15 representative blockfaces (totaling 1.8 curb-km) in the pilot study area (240 curb-km). Sampling days and times were selected randomly to minimize biases, although approximately 75 percent of all collections were performed before noon. Street litter (collected from center of road to curb) was bagged separately from sidewalk litter (collected from curb to building line). Sampling protocol for each blockface involved two field crew members, each of whom collected all loose litter from either the street or the sidewalk using a broom, a longhandled sweep pan, and a wheeled garbage bin lined with a plastic bag. Natural materials (leaves, stick, etc.), gravel, sand, bricks, animal remains and droppings, and items pinned under parked vehicles were ignored. Tires, large appliances, and similar "bulk" items were not collected, but were noted on field log sheets. Hazardous items such as syringes and glass shards were retrieved with tongs and placed inside hard plastic containers for safe handling. Except for petroleum and other environmentally sensitive substances, liquids were drained from uncapped containers prior to collection. After sweeping each blockface, each sample bag was tagged with an identification label and subsequently returned to a field laboratory at the end of the day. At the laboratory, technicians first weighed the unopened sample bag and then emptied the bag onto a table for sorting. Inadvertently collected nonlitter material, such as leaves and dirt, were discarded. Litter items were sorted into 13 material categories (plastic, paper, etc.) and arranged on the table in rectangular areas, with items of like material laid flat adjacent to the next piece. Regardless of material makeup, items which sank when placed into a test basin of water were sorted together as a separate "nonfloatable litter" category. After all items in the sample bag were sorted, the total silhouetted "surface area" occupied by each category was measured using a grid pattern printed on the table. Item counts and total weights were also measured for each category. These characteristics and other observations such as sample wetness were recorded on a laboratory log sheet. The samples were then placed into a 75 liter (20 gallon) garbage can which, when full, was subsequently used to determine the loosepacked density of each material type. In addition, about 1 in 10 sample bags were subject to a characterization of individual items according to a 47 item type classification used in the CityWide Floatables Study (HydroQual, 1993). These litter characterizations were conducted to allow comparisons between the collected street floatables and the washup material found on beaches, and to quantify the collected material in terms of bulk as well as visual impact. Item counts alone cannot fully describe the visual impact of floatable litter because the size of each item is not represented (a cigarette butt is equivalent to a large plastic sheet). Similarly, material weights can skew floatable measures since items of similar size can have much different weights (e.g. a glass bottle versus a plastic bottle). Finally, surface areas and item counts do not measure the "bulk" quantities of volume and weight important for waste disposal considerations. The results of these floatable litter characterizations are not within the scope of this paper and are described in greater detail elsewhere (HydroQual, 1995). Pilot Study Results As measured by the field monitoring program, the results of the pilot study clearly show that manual sweeping practices provided significant reductions in the quantity of litter present on City streets. Level 1 enhanced cleaning reduced street litter levels from Baseline conditions, and Level 2 enhanced cleaning generally reduced litter levels from Level 1 conditions. As presented in the following discussion, increasing the street cleaning frequency from two to six sweeps per week (Baseline to Level 1) provided a much greater marginal benefit than increasing cleaning from six to 12 sweeps per week (Level 1 to Level 2). Effects of enhanced street cleaning were, as expected, less pronounced on the sidewalks than on streets. While all streets benefited from Level 1 enhanced cleaning, the most littered streets did not benefit further from the doubled effort of Level 2 cleaning. The effect of enhanced cleaning on streets in certain landuse categories was not clear, but indications are that low density residential areas did not benefit from enhanced sweeping as much as other areas, and that streets along vacant lots benefited the most. The average quantities of floatables found on the monitored blockfaces in the study area are summarized for each of the three cleaning conditions in Table 2. All measures (item counts, surface areas and weights) are normalized per 100 meters of curb length, which is approximately equal to the length of a typical city blockface. Item count averages of all discrete street litter samples indicate the highest value (72.5 floatable items per 100 meters of curb length) during the Baseline condition, when streets were cleaned twice a week, a lower value (42.0 items/100 m) during the Level 1 condition, when streets were cleaned six times a week, and the lowest value (35.1 items/100 m) during the Level 2 condition, when streets were cleaned 12 times a week. The same trend is evident in the surface area and weight measurements of the collected street floatables. Increasing cleaning efforts also reduced the quantities of floatable litter on the adjacent sidewalks (Table 2, bottom). For all cleaning conditions, more floatable litter was present on the sidewalks than on the streets.
The effectiveness of each level of enhanced cleaning may be evaluated by comparing the incremental reductions between the average quantity of floatables observed during each cleaning condition. These reductions are presented in Table 3 as percentage changes with respect to the Baseline quantities. For street floatables, the average Baseline item count (72.5) minus the average Level 1 count (42.0) yields a reduction of 30.5 items per 100 meters, or a 42 percent reduction in the Baseline value due to the added Level 1 manual sweeping. The incremental difference between Level 1 and Level 2 of 6.9 items per 100 meters (42.035.1) indicates an additional reduction of 10 percent in the Baseline item count.
For street samples, the incremental reductions between Baseline and Level 1 (42 percent for item counts, 51 percent for surface areas, and 64 percent for weights) are greater than the reductions between Level 1 and Level 2 (10, 6 and 5 percent for item counts, surface areas and weights, respectively). Thus, the reductions in floatables diminish with the incremental cleaning efforts involved, i.e., the additional four sweepings between Baseline and Level 1, and the additional six sweepings between Level 1 and Level 2 (see Table 1). On a "per sweep basis," each Level 1 manual sweep removes an average of 7.6 additional floatable items per 100 meters, while each of the six additional manual sweeps conducted under Level 2 removes only 1.1 additional floatable items per 100 meters. Overall, each of the 10 Level 2 manual sweeps removes 3.7 additional items per 100 meters beyond Baseline cleaning. Reductions in floatable litter quantities are less pronounced on sidewalks than on the streets. This was expected, since the manual sweeping performed under Levels 1 and 2 concentrated on removing litter from the streets. Personnel performing manual sweeping did not step off of the street and only removed sidewalk litter from the portion of the sidewalk that they could reach with their brooms from the curb. Sidewalk litter sampling, however, was performed for the entire width of the sidewalk, from the curb to the building or fence line. Since the additional manual cleaning affected only a small portion of the sidewalk area, the quantity of floatable litter collected from the sidewalks was not substantially reduced during the enhanced cleaning program. Other Considerations Results of the NYCDOS enhanced cleaning pilot study are herein evaluated through comparison of the quantities of floatable litter observed during each cleaning condition. However, other factors besides street cleaning can affect the amount of litter on City streets. These factors include litter deposition rates, blockface land-use, rainfall, and on-street motor vehicle parking. Effect of Litter Deposition Rate - In order to determine whether the enhanced cleaning efforts were more (or less) effective on more heavily littered blockfaces than on other blockfaces, analyses were conducted to examine sweeping effectiveness for blockfaces with similar Baseline litter levels. In this context, the Baseline litter levels indicate relative litter deposition rates, which are assumed to remain constant for the duration of the study period. Note that although landuses (such as the presence of carryout food businesses) remained unchanged during the study period, the effect of the enhanced sweeping program on public behavior was unknown, and this analysis therefore assumes that any changes in litter deposition rates occurred uniformly within the study area. Figure 2 illustrates the effects of the enhanced cleaning programs on each of three groups of blockfaces, as categorized by Baseline street cleanliness. Two blockfaces considered cleanest of all, with street floatable item counts of fewer than 30 per 100 meters during Baseline cleaning, exhibit further reductions in litter under Level 1 and Level 2 enhanced cleaning (Figure 2, left). Nine blockfaces, considered "acceptably clean" according a visual rating system administered by the New York City Mayor's Office of Operations, also show reductions in litter between Baseline and Level 1, and between Level 1 and Level 2, although the incremental reductions decrease with increasing cleaning effort (Figure 2, center). The four most highly littered blockfaces, with Baseline floatable litter counts of 120 items per 100 meters, show great improvement under Level 1 cleaning but no further improvement under Level 2 cleaning.
Incremental reduction percentages between the various cleaning levels are summarized in Table 4 for the three blockface cleanliness levels, as measured by item counts, surface areas and weights of street floatables. Regardless of blockface cleanliness, the incremental reductions from Baseline to Level 1 are much higher than from Level 1 to Level 2 as noted previously in Table 3. While the "cleanest" and "acceptably clean" categories do show some additional litter reductions between Level 1 and Level 2, the extra daily manual sweeping was ineffective in removing additional floatable litter from the four most highly littered streets.
Effect of LandUse Category - The effectiveness of enhanced street sweeping was also examined with regard to the landuse of the blockfaces. Each of the 15 monitored blockfaces in the study area was categorized into the following landuse groups: residentiallow density (one and two family homes), residentialhigh density (multiple family dwellings), commercial, industrial, vacant and parking lots, and parks and institutions. Although each blockface was generally comprised of several landuses, one category was assigned if that landuse accounted for at least 50 percent of the total frontage of the blockface. Two blockfaces had no predominant landuse and were deleted from this analysis. It is important to note that, due to the small number of blockfaces in each landuse category, results of this analysis should be interpreted judiciously. As shown on Figure 3, the two blockfaces classified as low density residential have an average count of 50 floatable items per 100 meters of street under Baseline cleaning, a nearly equal number under Level 1 cleaning, and about half that number under Level 2 cleaning. The three high density residential blockfaces show a large reduction of litter under Level 1 cleaning and a much smaller incremental reduction under Level 2 cleaning. Essentially no incremental improvement in litter levels is shown between Level 1 and Level 2 cleaning on streets with commercial (four blockfaces), industrial (two blockfaces), and parks/institutions (one blockface) landuse categories. The vacant/parking lots blockface had the greatest Baseline floatables count (141 items per 100 meters) and the greatest incremental reduction in floatable litter between Baseline and Level 1 cleaning.
Incremental percentage reductions are summarized by landuse category in Table 5 for floatable item counts, surface areas and weights. The one street classified as vacant/parking lots had a 72 percent reduction in floatable items, a 66 percent reduction in surface area, and an 83 percent reduction in weight from Baseline to Level 1 ranking it among the most affected landuse types. Across all landuse categories except the low density residential category, the incremental effect of Level 1 cleaning was much greater than the incremental effect of Level 2 cleaning. For the low density residential category, there was virtually no difference between Baseline and Level 1 cleaning with respect to item counts and surface areas, although weightbased calculations were more in line with the trends seen for other landuse classes. Again it must be emphasized that, due to the limited number of blockfaces in each landuse category, trends associated with any particular landuse should be interpreted judiciously.
Effect of Rainfall - Rainfall can act to wash floatable litter from streets and sidewalks. If different rainfall conditions occurred during the various Baseline and enhanced cleaning periods, it is possible that the effects of street sweeping could be masked by the flushing action of rainfall and rainfall runoff. The relative importance of rainfall characteristics, such as number of storms, time between storms, maximum storm intensity and total storm volume, are somewhat unclear with respect to their effect on street cleanliness. Examination of rainfall records near the study area (Table 6) reveals that the number of rainfall events each month was essentially the same for the Baseline and Level 2 conditions, whereas a smaller number of events occurred during the Level 1 period. Therefore, the incrementally greater effect associated with Level 1 per month cleaning cannot be attributable to the number of rainfall events. Furthermore, the highest average monthly rainfall fell during the Baseline period, when street floatables levels were highest. However, the highest rainfall intensities were observed during the Level 1 period, possibly indicating that the flushing rates were higher during Level 1 storms.
Effect of On-Street Vehicle Parking - An additional factor, which may have had some influence on the lower incremental floatable litter Parking (ASP). ASP signs removals observed for Level 2 cleaning, is the practice of Alternate Side of Street restrict parking to one side of of the street two days each the street on certain days; typically, cars may not park on a given side week. This practice allows mechanical twice a week without having to street sweeper trucks to drive along each side of the street maneuver between parked cars, assuming no cleaning, manual sweeping was cars are illegally parked. Under Level 1 enhanced always where parked cars could obstruct performed on blockfaces where mechanical sweeping was not scheduled and, hence, the curb. Thus, cars could Level 1 cleaning. Level 2 sweeping obstruct the curb for 67 percent of the sweeps conducted during involved an of the Level 2 sweeps. Since most additional manual sweep each day, so cars could obstruct the curb for 83 percent street litter is found within about a half a meter from the curb, less efficient expected during Level 2 cleaning.removals would be CLEANLINESS OF AREAS WITHIN BUSINESS IMPROVEMENT DISTRICTS In addition to evaluating the effectiveness of the enhanced cleaning programs that the NYCDOS administered in the pilot study area, additional monitoring was performed on blockfaces cleaned by private contractors in self-taxing Business Improvement Districts (BIDs). As part of an overall city-wide litter assessment conducted during the summer of 1993 (HydroQual, 1995), 12 commercial blockfaces were monitored for street and sidewalk litter levels. All 12 of these blockfaces are cleaned regularly under the existing NYCDOS mechanical street sweeping program. Six of the blockfaces are located within three different BIDs, each of which augment the NYCDOS street cleaning services with manual sweeping of streets and sidewalks as many as 8 to 12 times each day, 7 days a week. Since litter levels on the BID blockfaces could not be monitored without the activities of the BID sweepers, this evaluation of the effect of BID sweeping uses the non-BID blockfaces as a "control" group. Comparisons of litter levels monitored on these BID and non-BID commercial blockfaces clearly indicate that employment of private contractors to clean streets and sidewalks can reduce street litter levels significantly. As shown in Table 7, BID blockfaces receiving both normal NYCDOS mechanical street sweeping and nearly continuous privately contracted street and sidewalk sweeping averaged 76 percent fewer floatable items, 73 percent less floatable surface area, and 82 percent less floatable weight as compared to similar commercial streets receiving only the NYCDOS twice weekly mechanical street sweeping. Assuming that BIDs manually sweep streets an average of 70 times each week, on a "per sweep" basis each BID sweeping removed an additional 0.60 items per 100 meters from the streets,([56.1-13.8]/70).
COSTS OF STREET CLEANING Information was obtained from NYCDOS and BIDs to develop costs for various street cleaning programs. A summary of these cost estimates is presented in Table 8, in terms of annual dollars per swept curb-kilometer.
The total cost associated with current NYCDOS street cleaning activities is complicated by the fact that NYCDOS also handles many other sanitation services, from street cleaning after special events to refuse collection, recycling, and snow removal. Peripheral agencies and activities, and the overhead associated with them, also contribute to the net overall cost of cleaning programs. That being said, rough estimates of the cost of cleaning the approximately 18,800 swept curb-kilometers in New York City with mechanical sweepers were developed on an annual cost per curb-kilometer basis. Considering only the capital and operating costs of the mechanical sweepers, the salaries and benefits of maintenance staff, salaries of drivers, and waste disposal costs, NYCDOS spent roughly $40 million in fiscal year 1993 (approximately $2,000 per curb kilometer). These numbers would increase by approximately 25 percent to account for overhead at the City government level. Based on the costs of the Enhanced Sweeping Pilot Study, NYCDOS developed incremental cost estimates for implementation of Level 1 and Level 2 enhanced sweeping practices in virtually all areas of the City currently swept mechanically. The annual cost for Level 1 cleaning (manual sweeping each blockface once per day, four days per week) was estimated to be $5,800 per curb-kilometer. The cost for Level 2 cleaning (manual sweeping each blockface once per day two days per week and twice per day four days per week) was estimated to be $14,000 per curb-kilometer. These incremental costs do not include the cost of existing mechanical sweeping nor any overhead costs associated with mechanical or manual sweeping. For comparison, street cleaning costs were estimated for the three BIDs within which six survey blockfaces had been monitored as part of this study. Each of these BIDs operates in midtown Manhattan, performs manual cleaning 7 days a week, and sweeps each blockface as many as 8 to 12 times each day. In addition to street cleaning, however, sanitation services provided by these BIDs typically include emptying litter baskets as well as sweeping. The associated level of effort translates to roughly 90 to 230 man-hours per week per curb-kilometer, and costs roughly $55,000 to $80,000 per curb-kilometer per year, exclusive of any management overhead. COST EFFECTIVENESS OF STREET CLEANING As discussed above, the various manual enhanced cleaning programs monitored in this study reduce levels of floatable litter on NYC streets. However, annual costs of these labor intensive programs are high. The relative cost benefits of Level 1, Level 2 and BID sweeping programs may be compared through an assessment of the incremental costs and floatable street litter reductions of each program with respect to the Baseline conditions. Figure 4 presents a "knee of the curve" analysis of the cost benefits of each program. Note that the percentage litter reductions shown reflect measures by item counts, surface areas and weights for each cleaning (cost) level. The small incremental reductions obtained by increasing the number of manual sweeps beyond four per week (Level 1) appear to be uneconomical.
With respect to other alternatives for floatables control, street sweeping offers various advantages and disadvantages. Besides being an alternative for control of floatables in waterways and beach washups, street sweeping helps to beautify urban areas, improve the quality of life and provide employment opportunities. Manual street sweeping does not require the often extensive capital costs and time delays often associated with other engineering alternatives, and can often be instituted in a short period of time. In addition, manual street sweeping is flexible and can be performed in almost any area of the City, since there are virtually no land or energy requirements. On the other hand, manual street sweeping is expensive. A rough comparison of the cost of manual street sweeping compared to other engineering alternatives (Table 9) shows that the other methods, where practical, are more cost effective in reducing floatable litter loadings to area waterways. Therefore, the expense of manual street sweeping limits its use to temporary periods or to instances where other alternatives are not practical.
CONCLUSIONS
ACKNOWLEDGMENTS Excerpts of this paper appeared in the October 1997 issue of Watershed & Wet Weather Technical Bulletin published by the Water Environment Federation. The work described within this paper was conducted by HydroQual, Inc. for the City of New York, Department of Environmental Protection, Bureau of Environmental Engineering. Authors:
REFERENCES HydroQual, Inc. (1993). Sources, Fate and Control of Floatable Materials in New York Harbor. Final Report, City-Wide Floatables Study for New York City Department of Environmental Protection, Bureau of Environmental Engineering, Division of Water Quality Improvement; December 1993. HydroQual, Inc. (1995). Evaluation of Non-Structural Methods to Control Combined and Storm Sewer Floatable Materials. Final Report, Floatables Pilot Program, City-Wide Floatable Study for New York City Department of Environmental Protection, Bureau of Environmental Engineering, Division of Water Quality Improvement; January 1995 (finalized December 1995). |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Return to HydroQual Presentations | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
