Pedestrian and Bicycle Safety in the CNVR: ý ó An Assessment of Existing Conditions Council of Governments of the Central Naugatuck Valley Approved by the Council of Governments: February 11, 2010 TITLE: Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions AUTHOR: Council of Governments of the Central Naugatuck Valley (COGCNV) SUBJECT: Study of existing conditions and hazards for bicycles and pedestrians. DATE: December 2009 REGIONAL PLANNING AGENCY: Council of Governments of the Central Naugatuck Valley SOURCE OF COPIES: Council of Governme nts of the Central Naugatuck Valley 60 North Main Street, 3 rd Floor Waterbury, CT 06790 NUMBER OF PAGES: 27 ABSTRACT: This study compiles data on automobile collisions involving pedestrians and bicycles in th e Central Naugatuck Valley Region (CNVR) and analyzes the spatial distribution and characteristics involved in these accidents. Th e study includes recommendations to address bicycle and pedestrian safety issues in the region. * * * * * * The preparation of this report was financed through grants from the U.S. Department of Transportation, Federal Highway Administration a nd the Federal Transit Administration; a grant from the Connecticut Department of Transporta tion; a grant from the Connecticut Office of Policy and Management; and by contributions from member municipalities of the Central Naugatuck Valley Region. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions CONTENTS I. Introduction Methodology……………………………………… ………………………….1 Geospatial Analysis………………………………………… ……………….1 Scope of State and National Data .………………………… ……………….2 II. Existing System Facilities for Pedestrian & Bicycles……………..………… ……………..…3 III. CNVR Pedestrian and Bicycle Accident Locations Pedestrians………………………………………… ……….…………….…5 Bicycles………………………………………… ……….………………….9 . IV. Analysis of Surrounding Land Use and Socio-Economic Characteristics Pedestrians………………………………………… ……….………………13 Bicycles………………………………………… ……….…………………18 V. Recommendations…………………………………… ………….………………22 VI. Public Comments …………………………………………… ….………………25 Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions I. Introduction This study compiles data on automobile collisions involving pedestrians and bicycles in the Central Naugatuck Valley Region (CNVR) a nd analyzes the spatial distribution and characteristics involved in th ese accidents. The purpose of the study is to identify common factors in reported bicycle and pedest rian accidents and locations that should be considered for future study. One objective of COGCNV’s regional transportation plan is to “to increase the safety and security of the transportation system for motorized and non- motorized users ” as part of our larger goal of de veloping and maintaining “an efficient transportation system that will provide the public with a high level of mobility, safety, and choice, while also addressing social , economic, and environmental needs and concerns.” 1 Methodology The study uses accident data provided by the Connecticut Department of Transportation (ConnDOT) for all vehicle collisions involving pedestrians and bicycles from 2003 to 2007. 2 Incidents labeled as “Pedestrian” accide nts in police reports do not adequately capture the whole universe of data, and “Bicyc le” accidents are not classified. The data still underestimates th e total number of cases, since it misses those accidents that go unreported. Time periods, injuries, and contributing factors in reporte d accidents are also analyzed in the study. The initial statistical and spatial analysis of collision data allows us to identify preliminary relationships between accident frequency and socio-economic and land use characteristics. Statistical analysis was pe rformed for the aggregate data of the whole region, the City of Waterbury, and the twelve towns that constitute the rest of the CNVR. Accidents involving pedestrian s and bicycles were geo-coded and mapped to help identify any correlation with socio-ec onomic characteristics and land use. Geospatial Analysis Accidents were geo-coded, mapped, and aggr egated to the census block group level for spatial analysis. Point data representing each accident was overlaid on data layers representing socio-economic and land use char acteristics to determine whether there is any correlation between the incidents and th e presence of certain socioeconomic, land use, and roadway variables, using Geogr aphic Information System (GIS) mapping. A map of accident densities was created with ArcGIS Spatial Analyst software to identify hazardous locations with high frequencies of pedestrian and bicycle collisions. Further studies of these locations can identify the appropriate safety improvements. Since there is little data available for pedestri an and bicycle traffic in the region, these maps can also be 1 COGCNV, Long Range Transportation Plan 2007-2035. July 2007. pp. 4-5 2 This includes “property-damage only” collisions in 2007. These collision types were not recorded for ConnDOT’s records prior to January 2007. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 2 used to identify areas where counts should be performed. The following characteristics were analyzed in this report.  Socio-economic Analysis  Poverty ─ Areas where 25% of the population is below the poverty level, using block group data from the 2000 Census. For perspective, during the 2000 Census, only 12% of the national population was belo w the poverty level, 8% of the state population, and 11% of the Wa terbury Urbanized Area.  Access to a vehicle ─ Areas where 25% of the population lacks access to a vehicle, using block group data from the 2000 Census.  Age ─ ConnDOT, CT Accident Summary Tables (CAST) (2007)  Spatial Analysis  Density of Accidents ─ ArcGIS Spatial Analyst was used to plot the density of “points” that represent accident locations  Accidents at Intersections  Land Use ─ COGCNV Geographic Information System (GIS) mapping data based on zoning and visual assessments o Regional Core ─ COGCNV Geographic Informa tion System (GIS) mapping data based on zoning, population dens ity, land use, and visual assessments o Community Centers ─ COGCNV Geographic Information System (GIS) mapping data based on zoning, populat ion density, land use, and visual assessments  Accident Characteristics  Hour  Day  Month  Contributing Factors  Injuries Scope of State and National Data All trends at the state and national leve l are drawn from 2007 data. References to statewide trends are drawn from summary tables produced by ConnDOT’s accident summary table database. In 2007, there were 1,277 motor vehicle accidents involving pedestrians and 820 involving bicy clists in the Connecticut. National trends were identified using the National Highway Traffic Safety Administration’s “Fatality Analysis Repor ting System” (FARS). This comprehensive resource is available to the public a nd may be accessed at www.fars.nhtsa.dot.gov. In drawing comparisons, however, it must be noted that data from FARS reflects only fatal accidents. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 3 II. Existing System The existing pedestrian and bicycle transportation facilities in the state focus on recreational uses. With the ex ception of New Haven, which has a rather large percent of walking commuters, 3 few of the state’s commuters use nonmotorized means of getting to work. According to the 2000 Census, 1.8% of the region’s workforce either walked or biked to work. The 2006-2008 American Comm unity Survey, which the Census Bureau uses to collect data annuall y, confirms that 2.3% of workers in Waterbury (~1,000 people) walk to work, while only a ve ry small number ride their bicycles. People, who walk and bike in the region, as in the rest of the state, are typically making short trips in high-density residential areas and central business districts or are using recreational paths. The first group experiences a greater safety concern as they must interact with motor vehicles regularly. While no data exists showing pedestrian or bicycle traffic volumes, the location of accidents found in the study pinpoints areas that should be targeted for future volume counts. Facilities for Pedestrian & Bicycles There are more than 60 off-road, multi-us e trails in Connecticut, including the Middlebury Trolley Line trail, the Larkin State Park Trail, and the Farmington Canal Trail in the Central Naugatuc k Valley Region (CNVR). Other trails are being planned in the state, primarily funded by the Federa l Highway Administration (FHWA) under the Transportation Enhancement Program. Along with the network of sidewalks located in the region’s urban and community centers, th e region’s multi-use trails are the main facilities serving recreational users. Municipalities in the Naugatuck Valley are planning a greenway along the Naugatuck River with some already establishing sections that will be incorporated into the finished route. Within the region, COGCNV is devel oping a regional Naugatuck River greenway routing study, and Waterbury is completing a routing study for the City’s portion. The CNVR portion was designated an official state greenway in 2003. A number of other recreational paths, ca lled Blue-Blazed Hiking Trails, ar e located in the region. The CT Forest & Parks Association publishes the CT Walk Book West as a resource for locating the trails. In its Bike Plan: 1994 , COGCNV recommended several state routes to be used as potential bike lanes, shared routes, and inter-regional routes . See Figure 1 for a map of the proposed routes. In addition, the state is in the process of updating its Bicycle and Pedestrian Plan, which includes a map of bicycle routes that are rated based on their suitability. The map, which is still in draf t form as of October 2009, can be found at www.ctbikepedplan.org . Most of the CNVR routes in cluded on the state’s bicycle map are the same as those recomm ended in COGCNV’s 1994 plan. 3 According to the 2000 Census, New Haven has the fourth highest rate of walking commuters among cities with over 100,000 people. !!!!!!!! !!!! ! !!!! ! !!! ! !! ! !! !!! !!! ! !!!!!! !! !!! !!!!!!!!!!! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! !!! ! !!!!! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! !! !! ! ! !!! ! ! ! ! ! !! ! !!!!! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !!!!!! ! ! !!!!!!! !!! !!!!!!! ! ! ! ! !!! !! ! ! ! !!!! !! ! ! !!! ! !! ! !! !! !!!!! ! ! !! ! ! ! ! ! ! !! !! !! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! !! ! !! !! ! ! B R I D G E W A T E R ” ¥ ” ¥ ” § ” Â ” Í ” Î ” Ò ” Ñ ” Ó ” Å ” Ì ” × ” ð ” ½ ” × ” Ý ” Ü £ t ” ì ” ¬ ” e ” Í ” Í ” ½ ” Ð ” Ð ” Â ” Â ” ¥ ” Ì £ t t ” Ó ” e ” Ò ” ð ” Ò ” Ñ ” ½ § ¨ ¦84 § ¨ ¦84 § ¨ ¦691 ” Í ” Ñ ” Î M O N R O E H A M D E N M O R R I S B R I S T O L R O X B U R Y M E R I D E N B E T H A N Y S E Y M O U R N E W T O W N P L Y M O U T H W A S H I N G T O N L I T C H F I E L D W A L L I N G F O R D S O U T H I N G T O N Oxford Southbury Cheshire Woodbury Waterbury Watertown Wolcott Bethlehem Middlebury Prospect Naugatuck Thomaston Beacon Falls Figure 1. Proposed CNVR Regional Bikeway System, 1994 ³ 0 24 1 Miles Council of Governments Central Naugatuck Valley Source: “Bikeways”, Bike Plan: 1994, COGCNV. “Roads”, c1984-2007 Tele Atlas, Rel. 01/08 “Town Boundary”, DEP Bikeways ! ! Bike Lanes ! ! Shared Lanes ! ! ! Interregional Routes For general planning purposes only. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 5 III. CNVR Pedestrian and Bicycle Accident Locations Pedestrians During the study period, 565 accidents involved pede strians. 81% occurred in Waterbury. Table 1. Pedestrian Accidents in the CNVR: 2003-2007 by Municipality Pedestrian Percent Municipality Accidents of CNVR Beacon Falls 3 < 1% Bethlehem 0 ─ Cheshire 10 2% Middlebury 2 < 1% Naugatuck 26 5% Oxford 3 < 1% Prospect 8 1% Southbury 5 < 1% Thomaston 12 2% Waterbury 458 81% Watertown 23 4% Wolcott 8 1% Woodbury 7 1% CNVR 565 100% High-hazard locations are re presented on maps on the follo wing pages. Figure 2 shows areas where the highest frequencies of accidents occur in the region. Figure 3 shows high frequency accident locations in Waterbury. Table 2 lists these locations with their respective vehicle traffic volumes. Corridors th at have a high concentration of accidents tend to have volumes in excess of 5,000 vehicles per day. B R I D G E W A T E R " ¥ " ¥ " § " Â " Í " Î " Ò " Ñ " Ó " Å " Ì " × " ð " ½ " × " Ý " Ü £ t " ì " ¬ " e " Í " Í " ½ " Ð " Ð " Â " Â " ¥ " Ì £ t t " Ó " e " Ò " ð " Ò " Ñ " ½ § ¨ ¦84 § ¨ ¦84 § ¨ ¦691 " Í " Ñ " Î M O N R O E H A M D E N M O R R I S B R I S T O L R O X B U R Y M E R I D E N B E T H A N Y S E Y M O U R N E W T O W N P L Y M O U T H W A S H I N G T O N L I T C H F I E L D W A L L I N G F O R D S O U T H I N G T O N Oxford Southbury Cheshire Woodbury Waterbury Watertown Wolcott Bethlehem Middlebury Prospect Naugatuck Thomaston Beacon Falls Figure 2. Density of Pedestrian Accidents Central Naugatuck Valley Region: 2003-2007 ³ 0 2.5 5 1.25 Miles Council of Governments Central Naugatuck Valley Density of Pedestrian Accidents – per square mile m LOW HIGH 4 246 Source: Accident Density was calculated by COGCNV using Spatial Analyst's Density Tool (Kernel Density) "Accident Data", ConnDOT Accident Record Section, 2003-2007. "Roads", c1984-2007 Tele Atlas, Rel. 01/08 "Town Boundary", DEP For general planning purposes only. " ¥ " Ò § ¨ ¦84 MIDDLEBURY NAUGATUCK PROSPECT CHESHIRE WOLCOTT WATERTOWN THOMASTON PLYMOUTH " ¥ ³ 0 1 2 0.5 Miles Source: Accident Density was calculated by COGCNV using Spatial Analyst's Density Tool (Kernel Density) "Accident Data", ConnDOT Accident Record Section, 2003-2007. "Roads", c1984-2007 Tele Atlas, Rel. 01/08 "Town Boundary", DEP Council of Governments Central Naugatuck Valley Density of Pedestrian Accidents – per square mile m LOWHIGH 4 477 Figure 3. Density of Pedestrian Accidents Waterbury: 2003-2007 For general planning purposes only. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 8 Table 2. High Hazard Locations for Pedestrians in the CNVR Geographic Area Average Daily Traffic (vehicles/day) Waterbury Downtown between Grand, Meadow, Grove, and N. Elm Street E. Main Street from the Green to Wolcott Road W. Main Street from the Green to Thomaston Avenue N. Main Street in the vicinity of East Farm Street S. Main Street in the vicinity of East & West Liberty Street Willow Street in the vicinity of Ridgewood Street While not as severe, the following locations exhibit high accident frequencies in the rest of the region. Beacon Falls N. Main Street from Route 42 to Church Street Cheshire South Main Street in the vicinity of Highland Avenue (Route 10) Naugatuck Meadow Street from Hillside to Rubber Avenue Rubber Avenue from Meadow to Aetna Street Maple Street from High to Church Street Prospect Route 68 in the vicinity of Route 69 Watertown Main Street (Route 63) from Route 6 to Woodruff Avenue Main Street, Oakville, (Route 73) from Davis to Buckingham Street Thomaston Route 6 in the vicinity of Route 109 Main Street from Route 254 to E. Main Street Woodbury Main Street (U.S. Route. 6) from Middle Quarter to Sherman Hill Road 4,000-8,000 15,300 22,800 (max.) 13,200 6,400 7,600 10,500 27,900 (max.) 8,100 16,100 12,600 11,300 (max.) 20,100 (max.) 15,300 (max.) 10,700 12,300 (max.) 16,800 (max.) Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 9 Bicycles During the study period, 185 accidents involved bicycles. About two-thirds of these accidents occurred in Waterbury. Among suburban towns, Naugatuck had the most with 9% of the re gional total. Table 3. Bicycle Accide nts in the CNVR: 2003-2007 by Municipality Bicycle Percent Municipality Accidents of CNVR Beacon Falls 2 1% Bethlehem 0 0% Cheshire 7 4% Middlebury 2 1% Naugatuck 17 9% Oxford 2 1% Prospect 5 3% Southbury 4 2% Thomaston 6 3% Waterbury 125 68% Watertown 7 4% Wolcott 5 3% Woodbury 3 2% CNVR 185 100% High-hazard locations are re presented on maps on the follo wing pages. Figure 4 shows areas with the highest frequencies of acci dents in the region. Figure 5 shows high frequency accident locations in Waterbury. Table 4 lists these locations with their respective vehicle traffic volumes. Corridors th at have a high concentration of accidents tend to have volumes in excess of 5,000 vehicles per day. B R I D G E W A T E R " ¥ " ¥ " § " Â " Í " Î " Ò " Ñ " Ó " Å " Ì " × " ð " ½ " × " Ý " Ü £ t " ì " ¬ " e " Í " Í " ½ " Ð " Ð " Â " Â " ¥ " Ì £ t t " Ó " e " Ò " ð " Ò " Ñ " ½ § ¨ ¦84 § ¨ ¦84 § ¨ ¦691 " Í " Ñ " Î M O N R O E H A M D E N M O R R I S B R I S T O L R O X B U R Y M E R I D E N B E T H A N Y S E Y M O U R N E W T O W N P L Y M O U T H W A S H I N G T O N L I T C H F I E L D W A L L I N G F O R D S O U T H I N G T O N Oxford Southbury Cheshire Woodbury Waterbury Watertown Wolcott Bethlehem Middlebury Prospect Naugatuck Thomaston Beacon Falls ³ 0 24 1 Miles Council of Governments Central Naugatuck Valley Source: Accident Density was calculated by COGCNV using Spatial Analyst's Density Tool (Kernel Density) "Accident Data", ConnDOT Accident Record Section, 2003-2007. "Roads", c1984-2007 Tele Atlas, Rel. 01/08 "Town Boundary", DEP Density of Bicycle Accidents – per square mile m LOW HIGH 4 60 Figure 4. Density of Bicycle Accidents Central Naugatuck Valley Region: 2003-2007 For general planning purposes only. " ¥ " Ò § ¨ ¦84 MIDDLEBURY NAUGATUCK PROSPECT CHESHIRE WOLCOTT WATERTOWN THOMASTON PLYMOUTH " ¥ ³ 0 12 0.5 Miles Source: Accident Density was calculated by COGCNV using Spatial Analyst's Density Tool (Kernel Density) "Accident Data", ConnDOT Accident Record Section, 2003-2007. "Roads", c1984-2007 Tele Atlas, Rel. 01/08 "Town Boundary", DEP Density of Bicycle Accidents – per square mile m LOWHIGH 4 177 Council of Governments Central Naugatuck Valley Figure 5. Density of Bicycle Accidents Waterbury: 2003-2007 For general planning purposes only. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 12 Table 4. High Hazard Locations for Bicyclists in the CNVR Geographic Area Average Daily Traffic (vehicles/day) Waterbury E. Main Street in the vicinity of Wolcott Street W. Main Street in the vicinity of Holmes Avenue N. Main Street in the vicinity of Division Street S. Main Street from East & West Dover to Washington Street Lounsbury Avenue in the vicinity of South Street Willow Street in the vicinity of Hillside Avenue Cherry Street from High to E. Main Street Walnut Street in the vicinity of Dikeman Street Bishop Street from Hawkins to Elizabeth Street Meadow Street at Freight Street While not as severe, the following locations exhibit high accident frequencies in the rest of the region. Cheshire Highland Avenue from Weeks Road to Cheshire High School Naugatuck Spring Street between Anderson Street and Route 68 15,300 22,200 13,200 7,800 550 7,600 11,500 6,900 unknown 13,200 27,900 (max.) 5,200 IV. Analysis of Surrounding Land Use and Socio-economic Characteristics Pedestrians Socio-economic characteristics Income ─ 49% of all accidents involving pedestrians were in areas where over one-quarter of the population earned below the poverty level (less than 1% of the region’s land area). 22% are within areas where over half the population is under the poverty level. Access to a Vehicle ─ 48% of all accidents involving pedestrians occurred in areas where over one-quarter of all households are without a vehicle (1.3% of the region’s land area). 20% were in areas where half of all households are without a vehicle. Land Use ─ 30% of all accidents involving pedestrians occurred in areas designated as Commercial, 44% as Residential – Medium density, and 7% occurred in areas designated as Residential – Low density. Approximately 85% of all accidents involving pedestrians occurred in the regional core and 5% occurred in community centers in the region. Figure 7 shows the distribution of bicycle and pedestrian accidents relative to the regional core and community centers. Time distribution Month ─ In looking at the monthly accident trends, there tended to be a higher frequency in the fall and early winter months. The majority of incidents in Waterbury occurred in November, while the rest of the region saw a spike in September and December. Waterbury also exhibited a trend from May through July, which it did not share with the rest of the region. Statewide, peaks were from September through December as well as May and June. The national trends show more of a focus in the fall and winter months with the highest frequency of pedestrian fatalities occurring in November. 13 Figure 6. Regional Distribution of Motor Vehicle- Pedestrian Accidents by Land Use: 2003-2007 3 0 % C o m m e rc ia l 4 3 .6 % R e s id e n tia l – M e d iu m D e n s ity R M C M R L C F T U R C R H U L I N O th e r 30% Commercial 43.6% Residential – Medium Density Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions B R I D G E W A T E R " ¥ " ¥ " § " Â " Í " Î " Ò " Ñ " Ó " Å " Ì " × " ð " ½ " × " Ý " Ü £ t " ì " ¬ " e " Í " Í " ½ " Ð " Ð " Â " Â " ¥ " Ì £ t t " Ó " e " Ò " ð " Ò " Ñ " ½ § ¨ ¦84 § ¨ ¦84 § ¨ ¦691 " Í " Ñ " Î M O N R O E H A M D E N M O R R I S B R I S T O L R O X B U R Y M E R I D E N B E T H A N Y S E Y M O U R N E W T O W N P L Y M O U T H W A S H I N G T O N L I T C H F I E L D W A L L I N G F O R D S O U T H I N G T O N Oxford Southbury Cheshire Woodbury Waterbury Watertown Wolcott Bethlehem Middlebury Prospect Naugatuck Thomaston Beacon Falls ³ 0 2 4 1 Miles Council of Governments Central Naugatuck Valley Source: "Accident Data", ConnDOT Accident Record Section, 2003-2007. "Roads", c1984-2007 Tele Atlas, Rel. 01/08 "Town Boundary", DEP "Regional Core" and "Community Centers", COGCNV Figure 7. Bicycle and Pedestrian Accidents in the Regional Core and Community Centers: 2003-2007 Bicycle Accidents Pedestrian Accidents Regional Core Community Centers cnv_bnd For general planning purposes only. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 15 Day ─ Accidents in the region commonly occur during the weekdays, primarily on Tuesdays and Fridays. In Waterbury, accidents are spread out on weekdays. The same is true at the state level, with most accidents occurring on Tuesdays. In the rest of the CNVR, however, the trend is focused on Fridays. This is more in line with national trends, which show that pedest rian fatalities are more common on weekend days (Friday – Sunday). Table 5 shows the most comm on days for pedestrian accidents. Hour ─ Accidents most commonly occur in th e evening between 3:00 and 8:00 p.m. at all geographic areas, with the highest frequencies from 4:00 – 5:00 PM. Table 5. Most Common Time Periods for Pedestrian Accidents in CNVR and CT Peak Time Periods Geographic Area Month Day Hour CNVR November Friday 4:00 p.m. Waterbury November Tuesday 4:00 p.m. Rest of CNVR September Friday 5:00 p.m. CT October Tuesday 5:00 p.m. Most of these trends can be attributed to pedestrians travelling during dusk and other seasonal factors. Snow banks, earlier sunset hours, and wearing darker clothing are all common in the winter months and play a combined role in causing the increase in pedestrian accidents. Other f actors that may be considered during the peak accident hours are increased traffic volumes during rush hour, school children walking home, and inadequate lighting on rural ro adways during sunset hours. Injury analysis There were 472 injury-causing accidents involving pedestrians during the study period, of which 23% included incapacitating injuries and 2% ( 11) were fatal injuries. Statewide, in 2007, there were 1,171 injury-causing accidents invol ving pedestrians, of which 20% included incapacitating injuries and 3% (32) fatal injuries. Both sets of statistics seem to indicate that the majority of accidents cause only minor injury. Intersections 62% of all accidents involving pedestri ans during the study period occurred between intersections (mid-block cro ssings). Of those incidents, 44% are attributed to unsafe use of highway by a pedestrian, which usually i ndicates that a crosswalk was not used. The locations are similar with the state trend, where 56% of all pedestrian accidents occurred between intersections, while the national inci dence reached a high of 77% of all fatal pedestrian accidents. The frequency of pedestrian accidents at mid-block crossings is indicative of children darting into the middle of the road and pedestrians seeking out the most convenient route to their destination. When destinations are located in areas without a crosswalk, motorists will be caught by surprise and may be unable to react in time. Snow banks, street parking, inadequate lighting, and other obstacles may further limit pedestrian visibility and conspicuity. )DFWRUVLQYROYHGLQUHSRUWHGDFFLGHQWV In the CNVR, accidents involving pedestrians are most often associated with “unsafe use of highway by pedestrians” and “failure to grant right-of-way.” These are the most common factors statewide as well. Nationally, the most common factors were “improper crossing of roadway or intersection,” and “walking/riding with or against traffic, playing, working, sitting, lying, standing, etc. in roadway.” Both would qualify under the state’s definition of “unsafe use of highway by pedestrian.” The two most common factors in the region can be associated with the party at-fault. When the vehicle is at-fault, it is often caused by the driver failing to grant right-of-way. The majority of these cases occur at unsignalized intersections, when a driver is turning and fails to stop for a pedestrian in the crosswalk. When the pedestrian is at-fault, on the other hand, it is typically classified as an unsafe use of the highway. This could mean that the pedestrian was either not using a crosswalk, hitching on a vehicle, or p laying in the road. It should be noted that the third most common contributing factor is “driver following too close” due to the number of rear-end collisions that were included in this study. Although no pedestrian was injured in these incidents, the accident records identify the role of a pedestrian in the collision, usually causing the first vehicle to stop or slow down abruptly. 16 Figure 8. Contributing Factors in Pedestrian Accidents in the CNVR: 2003-2007 Other Factors Unsafe Use of Highway by Pedestrian No Action Driving Too Fast for Conditions Driver Unable to Cope with Conditions Driver Following too Close Driver Failed to Grant Right of Way Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 17 Distribution by Age Data on pedestrian ages was only available for 2007, a year when there were 114 pedestrian accidents in the region. The data shows that the region’s pedestrian accidents occurred most commonly among people from 10 to 24 years old. In 2007, 39% of pedestrians involved in accidents in Connecticut and 43% of pedestrians injured nationwide were under 25 years old . While this trend is fairly uniform in the region, it is worth noting that 35% of pede strians involved in accidents outside of Waterbury were 45 and older. Table 6. Percent of Pedestrian Accidents by Age: 2007 Age CNVR Waterbury Rest of CNVR CT < 5 5% 6% 0% 3% 5-9 5% 6% 0% 5% 10-14 11% 11% 14% 8% 15-19 14% 13% 21% 12% 20-24 10% 9% 21% 8% 25-29 6% 6% 7% 6% 30-34 3% 3% 0% 4% 35-39 6% 7% 0% 6% 40-44 7% 8% 0% 7% 45-49 9% 9% 14% 7% 50-54 4% 4% 7% 6% > 55 13% 13% 14% 18% Unknown 8% 9% 0% 9% Total 100% 100% 100% 100% It is difficult to determine why some age groups are more susceptible to pedestrian accidents than others. Few studies have taken a comprehensive look at this topic, though some have noted that children are more likely to dart into the street, and young adults are more likely to cross at unmarked crossings, meaning that they are less careful about finding the safest place to cross. Children and older persons, on the other hand, are more likely to use marked crosswalks. 4 Pedestrian behavior varies based on a number of factors, including whether an area is urban or suburban, whether children are walking to school, and the presence of walkable destinations. 4 FHWA, Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations Final Report and Recommended Guidelines . http://www.tfhrc.gov/safety/pubs/04100/03.htm. Retrieved on 9/1/09. Bicycles Socio-economic characteristics Income ─ 35% of all accidents involving bicycles were in areas where over one-quarter of the population earned below the poverty level (less than 1% of the region’s land area). 13% are within areas where over half the population is under the poverty lev el. Access to a Vehicle ─ 36% of all accidents involving bicycles occurred in areas where over one-quarter of all households are without a vehicle (1.3% of the region’s land area). 12% occur in areas where half of all households are without a vehicle. Land Use ─ 25% of all accidents involving bicycles occurred in areas designated as Com- mercial, 46% as Residential – Medium density, and 13% occurred in areas designated as Residential – Low density. Approximately 75% of all accidents involving bicycles oc – curred in the regional core and 10% occurred in community centers in the region. Figure 7 shows the distribution of bicycle and pedestrian accidents relative to the regional core and community centers. Time distribution Month ─ The majority of bicycle accidents took place during the months of August and September in the region. This is likely because of the improved weather conditions and the associated increase in bicyclists on the roads. National and statewide trends were also higher in the summer months with peaks in the month of June and July, respectively. Day ─ Accidents commonly occur on Fridays and Tuesdays, followed closely by Satur- days. The majority of accidents statewide occur on Thursdays with a high frequency of ac- cidents spread throughout the weekdays. Table 7 shows the most common days for bicycle accidents. 18 Figure 9. Regional Distribution of Motor Vehicle-Bicycle Accidents by Land Use: 2003-2007 1 2 .4 % R e s id e n tia l – L o w D e n s ity 4 5 .4 % R e s id e n tia l – M e d iu m D e n s ity 2 4 .3 % C o m m e rc ia l R M C M R L T U I N C F R C O th e r 12.4% Residential – Low Density 24.3% Commercial 45.4% Residential – Medium Density Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 19 Hour ─ Accidents most commonly occur in the evening between 5:00 – 6:00 p.m. in the region. The statewide trend is similar with th e highest frequency of accidents occurring around 4:00 p.m. Table 7. Most Common Time Periods fo r Bicycle Accidents in CNVR and CT Time Period Geographic Area Month Day Hour CNVR September Friday 5:00 p.m. Waterbury September Saturday 5:00 p.m. Rest of CNVR September Friday 5:00 p.m. CT July Thursday 4:00 p.m. The rise in accidents during the late summer months likely coincides with the suitable weather and the beginning of the school season. As with the pedestrian accident statistics, it appears that the peak period coincides with evening rush hour traffic. However, there also appears to be less of a problem of acci dents occurring after sunset. While inadequate lighting can be hazardous for bicy clists, these figures show that most travel occurs during the daylight hours, which helps reduce the risk s associated with seasonal factors: snow banks, early sunset hours, etc. Injury analysis There were 170 injury-causing accidents among bicyclists, of which 31% included incapacitating injuries and 2% (3) fatal injuries. Statewide, in 2007, there were 657 injury-causing accidents among bicyclists, of which 33% included incapacitating injuries and less than 1% (5) were fatal injuries. As with the pedestrian accident data, it appears that the majority of bicycle-automobile accidents result in only minor injuries. Intersections 58% of all accidents involving bicycles occurred at intersections. Of those incidents, 44% are attributed to failure to grant right-of-way, which indicates a recurring problem at conflict points between motorists and bicycles at intersections. Statewide, 61% of pedestrian accidents occurred at intersections in 2007, but only 38% of all fatal accidents nationally occurred at intersections. While the regional data is in line with the state trend, there is a clear discrepanc y with the national trend. The discrepancy can be the related to a variet y of factors that are involved in accidents throughout the country. Other part s of the country may have a higher volume of cyclists travelling longer distances, who are more susceptible to collisions between intersections. Since the national statistics represent only fatal accidents, it is also possible that the sample is a reflection of the most dangerous accidents and a tendency for these to occur between intersections. Factors involved in reported accidents In the CNVR, among those accidents involving bicycles, the most common factor was “failure to grant right-of-way.” “Driving on the wrong side of the road” and “violating traffic controls” were tied for the second most common causes. The two most common factors cited in accidents involving bicycles statewide were “failure to grant right-of-way” and “violating traffic controls.” Nationally, the most common factor was “failure to grant right-of-way.” There is a clear trend in these collisions at all levels, where bicyclists and automobiles fail to grant the right-of-way or violate traffic controls, leading to collisions. These two factors show a definite need to improve education among bicyclists and auto- mobiles on how to properly “share the road.” Under state statute, bicyclists are granted the same rights on roadways as a motor vehicle. Motor vehicle operators should be mindful of their presence on the roadway, treat them as another vehicle, and provide sufficient clear- ance when passing so that they may use the road safely. Bicyclists are also subject to the same restrictions as automobile drivers, meaning that they are required to obey all of the rules of the road that automobiles are subject to. 20 Figure 10. Contributing Factors in Accidents Involving Bicycles in the CNVR: 2003-2007 Other Factors Driver on Wrong Side of the Road Driver Violated Traffic Control Driver Failed to Grant Right of Way Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 21 Distribution by Age Data on the age of bicyclists was only available for 2007. There were a total of 39 accidents involving bicycles in the region that year. Because of the way that accident data is compiled, the ages of bicyclists are combin ed with other drivers in accident summaries, since they are considered vehicle operato rs, unlike pedestrians who are classified separately. This makes it difficult to determin e the ages of bicyclists involved in these accidents. Of those individuals involved in bicy cle collisions, we can isolate those vehicle operators, who were under the age of 16 and therefore not allowed to operate automobiles. Within the state, that constitutes 15% of those involved in collisions, within the region 26%, within Waterbury 26%, an d within the rest of the region 18%. Nationally, that number is 29% of all bicyclists injured in collisions, although the average age is 30. The statistics show that th e problem is distributed am ong age groups with collisions among young people being a greater problem in urban areas. The trend among children aged 10 and younger is much more significant in Waterbury (22% ) than in the rest of the region, where the same age group are involved in less than 5% of all bicycle collisions. Table 8. Percent of Bicyclist Accidents by Age: 2007 Percent of Accidents, by age Age CT CNVR Waterbury Rest of CNVR 0-5 0% 1% 0% 0% 6-10 3% 17% 22% 4% 11-15 12% 8% 4% 14% 16 + 85% 74% 74% 82% Total 100% 100% 100% 100% Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 22 V. Recommendations The following recommendations are intended as guidelines for improving mobility and safety for nonmotorized users in the re gion. The recommendations are based on the spatial analysis and common characteristics in pedestrian and bicycle accidents identified in the report. The Federal Highway Administration (FHW A) has compiled a list of effective countermeasures in their PedSafe and BikeSafe manual 6 that can be used as guidelines in improving hazardous locations. These manuals provi de descriptions of useful treatments to common problems identified when looking at pedestrian and bicycle accidents. In the near-term, it would be useful to consider low-cost safety improvements as a means of reducing accidents in the most troubled locations, while planning to fix larger problems as long-term capital improvements. Near-term improvements should be directed towards areas with high accident frequencies that were identified on the accident density maps in Chapter III. Characteristics identified in Ch apter IV as exhibiting a higher frequency of accidents among bicyclists and pedestrian s should also be used to prioritize improvements and educational initiatives. Safety Improvements in High-Hazard Corridors and “Hot Spots” 1. Focus on high-hazard corridors identified in the study, combined with public input and anecdotal evidence from public works staff and traffic engineers to target high priority areas for improvements.  Establish safety impact teams made up of engineering professionals and local stakeholders.  Cooperatively develop improvement con cepts including design, education, and enforcement. Look at models like the NJDOT initiative, Pedestrian Safety Corridor Program, that targets corridors with a history of pedestrian safety problems.  Consult the Pedsafe and BikeSafe guides to determine appropriate countermeasures based on the accident da ta and characteristics of the troubled location.  For example, a high frequency of mi d-block pedestrian accidents may indicate areas in need of a crosswal k, areas where pedestrians walk in the roadway, faded crosswalk markings, or children darting in the road. The PedSafe guide recommends a number of treatments to reduce these accident types, including: o Providing a sidewalk on both sides of the road. o Improving nighttime lighting. o Installing medians or pede strian crossing islands. 6 FHWA’s “Pedestrian Safety Guide and Countermeasure Selection System,” September 2004. Available online at www.walkinginfo.org FHWA’s “Bicycle Countermeasure Selection System,” May 2006. Available online at www.bicyclinginfo.org Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 23 o Implementing traffic calming measures. o Installing warning signs to alert drivers of upcoming crosswalks.  For example, bicycle-automobile acciden ts that occur at intersections may indicate a concentration of conflict points. The BikeSafe guide recommends a number of treatments at these locati ons, including: o Improving lighting. o Enhancing pavement markings. o Signal optimization. o Making School Zone improvements. o Maintaining travel lanes and shoul ders free of debris and surface irregularities.  Prioritize corridors for funding. Enforcement 2. Plan enforcement campaigns and develop strategies to curb traffic violations by both motorized and non-motorized users. Education 3. Promote the use of reflective clothing a nd care when walking or riding along poorly- lit roadways to reduce pedestrian and bicycle accidents. 4. Educate motorists, bicyclists, and pedestri ans on traffic regulations that apply to them. Develop learning programs to teach safe and responsible bike riding and crossing techniques to school children. 5. Continue support for ConnDOT’s “Share the Road” campaign to alert drivers of the presence of cyclists and other non-motorized users. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions 24 Data Collection 6. Inventory crosswalks, pedestrian traffic generators (i.e. retail stores, schools), pedestrian sources (i.e. dense neighbor hoods, apartment buildings, bus stops) and “demand lines” (well-worn paths along the roadside).  Use GIS software to develop and implement sidewalk projects. Create a detailed inventory of sidewalks in high-traffic and high-hazard corridors to aid in prioritizing improvement pr ojects. Mapping sidewalks th roughout the region will also aid in planning for connectivity and increased pedestrian access. 7. Continue to accrue and monitor data to determine effectiveness of improvements. Identify areas where pedestrian and bi cycle traffic counts would be useful. Land Use and Transportation Design 8. Consider pedestrians and bicyclists in tran sportation design (i.e. complete streets, wide shoulders for cyclists). Assess the im pact of highway improvement projects on pedestrians and bicyclists. 9. Consider pedestrians and bicyclists in land use design (i.e. sidewalks in new developments). Land use decisions and development patterns should not hinder the mobility of non-motorized transportation users. “Demand line” along Thomaston Av enue (SR 847) in Waterbury. ConnDOT Highway Photolog: 2006. Pedestrian and Bicycle Safety in the CNVR: An Assessment of Existing Conditions VI. Public Comments Joseph Perrelli From: Lauren Rizzo [lrizzo@cogcnv.org] Sent: Tuesday, January 12, 2010 10:53 AM To: ‘Joe Perrelli’ Subject: FW: Draft Pedestrian & Bicycle Safety Study Attachments: Ped Bic Safety PDF.Final Draft.pdf Pa ge 1 of 1 1/12/2010 Lauren Rizzo Administrative Assistant COGCNV 60 North Main St., 3rd Floor Waterbury, CT 06702 203.757.0535 – phone 203.756.7688 – fax From: Lauren Rizzo [mailto:lrizzo@cogcnv.org] On Behalf Of Joe Perrelli Sent: Thursday, December 31, 2009 10:48 AM To: Linda Fercodini; Anthony Malone; Ellen Samoska; Geoffrey Green; Gil Graveline; Harmon Andrews; Harold Cosgrove; 'Herman Schuler'; James Sequin; Joseph McEvoy; Judy Wick ; Ken Long; Maria Hill; Martin Cobern; Martin Overton; Mary Barton; Ma ry Connolly ; Nancy Clark; Peter Betkoski ; Richard Minnick; Ruth Mulcahy; William Guerrera; 'Dan Norton '; 'David Kalinowski'; 'David Monckton'; 'Edward Bea'; 'James Stewart'; 'John Lawlor'; 'Paul Pronovos t'; 'Roy Cavanaugh'; 'Thomas Crowe'; 'Wayne Watt'; Chuck Berger; Denis Cuevas; Jim Galligan; Joseph Mi chelangelo; Mark Pronov ost ; Robert Oley; Scott Poryanda; Brian Miller ; Catherine Adsitt ; DeLoris Curtis; Jean Donegan; Michael Tanuis; Rebecca Au gur; William Donovan; William Voelker Cc: Stephen Livingston; David Balzer ; Peter Dorpalen; Virginia Mason Su bject: Dr aft Pedestrian & Bicycle Safety Study To:                     Regional Planning  Commission  (RPC),  Public  Works  Directors,  Town  Engineers,   and  Town  Planners From:                 Joe Perrelli,  Regional  Planner Subject:            Draft Pedestrian  &  Bicycle  Safety  Study Date:                 December  31,  2009         The  draft  Pedestrian  and Bicycle  Safety  in  the  CNVR:  An  Assessment  of Existing  Conditions  is   attached  to  this  email.   The study  assesses  existing  conditions  and  identifies  hazards  for bicy cles  and  pedestrians  in the  region.  It  was  presented  before the  Regional  Planning  Commission  (RPC)   at  the  September  meeting  and  will be  presented  for approval  at  the  next  RPC  meeting  on   January  12,  2010 . Please  review  the  study  and  provide  comments  prior to  the  meeting.   Written  comments  may  be  mailed  to  COGCNV  at  60  North  Main Street,  3 rd Floor,  Waterbury,   CT  06702,  or  e ‐mailed  to jperrelli@cogcnv.org .  If  you  have  any  questions,  please  contact  Joe   Perrelli  at  203.757.0535.    Joseph Perrelli From: 1DQF&ODUN Sent: Thursday, December 31, 2009 1:32 PM To: -RVHSK3HUUHOOL Subject: Re: Draft Pedestrian & Bicycle Safety Study Pa ge 1 of 1 1/12/2010 COG Bicycle Draft December 31, 2009   Joe:     Was there any data collected for “Frequency of Accidents in close proximity to Schools?”     Was there any data collected for “Purpose of Bicycle use?”   Ex:  pleasure / exercise vs  transportation to work / school?”     Also, I think that the urban vs rural   population affects the accident frequency due to number of cyclists, concentration of traffic,  and purpose of use.     This could factor into the conclusions for improvement of safety measures.     Nancy Clark, Southbury