Bicycle and Pedestrian Counts

The Delaware Valley Regional Planning Commission (DVRPC) launched its Bicycle Cyclical Count Program in 2014. Using the data collected over the last eight years, we can uncover trends in cycling behavior in Philadelphia and draw informed conclusions about what might be causing them.

DVRPC’s Bicycle Cyclical Count Program collects and analyzes bicycle volumes to track trends at 150 different locations in the nine-county region. These locations were selected for their representative attributes like having different types of bicycle facilities, such as multi-use trails, dedicated bicycle lanes, striped shoulders, sharrows, and in mixed traffic. DVRPC deploys the counters and collects data for seven days at each location every three years. The annual average daily bicycles (AADB) is then calculated each day by multiplying the raw count by the seasonal adjustment factor¹ and equipment factor². 

Based on 35 locations, between 2017 to 2019, Philadelphia experienced a seven percent net increase in AADB compared to 2014 to 2016. The growth was made up entirely of cyclists using dedicated bicycle lanes, including protected, buffered, and conventional. However, after the start of the COVID-19 pandemic, bicycle ridership in the City plummeted, and between 2020 to 2022, it was down 35 percent. Philadelphia's three most heavily ridden neighborhoods, including Central, South, and University-Southwest, steadily increased in AADB before the pandemic but experienced a significant decline afterward.

Trends by Neighborhood:

To better understand geographic factors impacting bicycle trends in Philadelphia, ridership was evaluated by neighborhood. In Figure 1, South Philadelphia, one of the three highest bicycle volumes neighborhoods in Philadelphia, had the highest increase of cyclists in the years before the pandemic, growing by 32 percent. It was significantly higher than Central and University-Southwest, which increased by nine percent and eight percent, respectively. However, between 2020 to 2022, bicycle ridership in South Philadelphia dropped more than 53 percent. In Central and University Southwest, bicycle ridership fell by 34 percent and 29 percent, respectively. In all three neighborhoods, ridership from 2020 to 2022 was below the level measured from 2014 to 2016, the beginning of data collection. 

Figure 1

Trends by Bicycle Facility Type: 

To understand cyclists' preferences and uncover behavior patterns, bicycle volumes were analyzed by facility type. As seen in Table 1, only dedicated bicycle lanes saw an increase in ridership among the measured bicycle facility types in the years before COVID-19. Protected and buffered bicycle lanes grew by 31 percent and 33 percent, respectively, while conventional experienced only a five percent increase, demonstrating a clear preference by cyclists. From 2020 to 2022, the AADB on all seven bicycle facility types was below the volume observed from 2014 to 2016. Dedicated bicycle lanes experienced the most significant decline in ridership over the last three years, with protected bicycle lanes dropping by 45 percent while buffered and conventional fell by 36 percent.

Table 1

Conclusion:

Since the beginning of DVRPC’s Bicycle Cyclical Count program in 2014,  bicycle infrastructure in Philadelphia has expanded to more city streets.  Notable improvements on the City’s bicycle network include the bollard and  parking-protected bicycle lane on Chestnut Street between 63rd Street to  34th Street in West Philadelphia, the two-way bollard and parking-protected  cycle track on 11th Street from Reed Street to Bainbridge Street in South  Philadelphia, and the bollard protected bicycle lane on 22nd Street from  Washington Avenue to Market Street in Center City.  

Indego, Philadelphia’s public bicycle share program, launched in 2015 with 600 bicycles at 60 stations³. The initial service area focused on Center City and nearby neighborhoods but has since expanded throughout the City. In 2015, 429,710 trips were completed on Indego; seven years later, in 2022, the number of completed trips more than doubled to 905,790⁴. Indego now has 200 stations with over 2,000 bicycles, consisting of conventional and electric bicycles, which were introduced in 2019. Indego plans to expand its network to 3,500 bicycles at 350 stations by the end of 2026⁵.

Even with infrastructure improvements, public initiatives, and policy changes from the City of Philadelphia, the COVID-19 pandemic has profoundly impacted the number of people riding bicycles, wiping out years of growth. Overall, cycling in the City grew by seven percent before the pandemic but decreased by 35 percent in the years since. The trend is evidenced by the decreased number of cyclists in almost every neighborhood and the reduced use of nearly every bicycle facility type, especially those heavily used before the pandemic. Outside of Philadelphia, this downward trend has been observed in peer cities. For example, in Portland, OR, bicycle ridership was also down 35 percent in 2022 compared to 2019 and decreased across every district in the City.

The decrease in cycling may partly be due to a rise in dangerous driving behavior by motorists. In 2021, eight cyclists died in Philadelphia, with seven killed by motorists, the highest number since data tracking began in 1997. In 2022, the number of cyclist fatalities fell to two in the City, but in the first couple of months of 2023, that number rose sharply, with four cyclists killed between January and March⁶. One trend in Philadelphia is that cyclist deaths are rising in the City even as fewer residents ride bicycles. Nationwide, there was an eight percent increase in cyclist fatalities between 2021 to 2022⁷. Traffic calming measures and multimodal infrastructure investments could create a safer environment for all road users. 

Bicycle volumes in Philadelphia have been significantly impacted by a change in commuting behavior as many office workers shifted to remote and hybrid work schedules during the COVID-19 pandemic. Nationwide, between 2019 to 2021, there was a significant decline in workers cycling to work, dropping by 24 percent⁸. In addition, the lingering impact of the pandemic can be observed in other modes of transportation; in April of 2023, SEPTA’s systemwide ridership was only 53 percent of its pre-pandemic levels, with its regional rail ridership at 54 percent and transit ridership at 53 percent⁹. Finally, the increase in working from home has led to a reduced occupancy rate in office buildings in Philadelphia. Per Kastle Systems, a security service firm that tracks entries into office buildings, the Delaware Valley’s occupancy rate is 48.9 percent, behind the national average of 57.7 percent¹⁰.

¹Correction factor to account for weather conditions, seasonal and weekday variations
²Correction factor to account for weather conditions, seasonal and weekday variations
³City of Portland - 2022 Bicycle Counts Report
⁴United States Department of Transportation - National Highway Traffic Safety Administration
⁵United States Department of Transportation - Bureau of Transportation Statistics
⁶Kastle Systems - Weekly Occupancy Report from Kastle Access Control System Data
⁷United States Department of Transportation - National Highway Traffic Safety Administration
⁸United States Department of Transportation - Bureau of Transportation Statistics
⁹SEPTA - April 2023 Passenger Revenue & Ridership
¹⁰Kastle Systems - Weekly Occupancy Report from Kastle Access Control System Data

The COVID-19 pandemic transformed nearly all aspects of society. This analysis examines its impact on pedestrian and bicycle traffic along two segments in the City of Philadelphia: Front and Chestnut Streets, which have an automatic pedestrian counter, and the Schuylkill River Trail near the Walnut Street Bridge, which has a pedestrian and bicycle counter. The two time periods evaluated were pre-pandemic, from March 2019 to March 2020, and July 2021 to July 2022, when some residents began resuming normal activities despite COVID-19’s lingering presence.

As seen in table 1 below, there has been a slow but steady increase of pedestrian traffic at Front and Chestnut relative to 2019 figures. Pedestrian volumes were down 7%, but this represents a significant uptick from 2021 when they were down as much as 40%. Along the Schuylkill River Trail, pedestrian traffic has returned to pre-pandemic levels but bicycle volumes were down 34%. In the first half of 2021, there was an abrupt decline in pedestrian and bicycle traffic as this section of the Trail was closed from February to July due to the reconstruction of the Chestnut Street Bridge (the data reviewed for this analysis is from before and after the closure). Additionally, MLK Drive, a popular bicycle connection to the Schuylkill River Trail that had been closed to motorists beginning in March of 2020, re-opened to car traffic north of Sweetbriar Drive in August of 2021. The re-opening of MLK Drive and the residual impact of the trail closure may have contributed to the decline of bicyclists on the Schuylkill River Trail.

Figure 1

As pedestrian traffic returned to Front and Chestnut, there was a significant shift in activity by time of day. As displayed in figures 2 and 3 below, there was a pronounced decrease in weekday pedestrian traffic during traditional lunch hours, typically from 11:30 am to 1:30 pm. This may be due to the ongoing prevalence of telecommuting as well as businesses reducing hours, such as restaurants pivoting to only offering dinner service due to labor shortages. Pedestrian traffic on Friday and Saturday evenings was still well below its 2019 levels but experienced a rebound, while Sunday volumes increased dramatically throughout the day, especially in the mid-afternoon.

Figure 2

Figure 3

On the Schuylkill River Trail, pedestrian traffic from July 2021 to July 2022 generally mirrors patterns seen before the COVID-19 pandemic, while bicycle volumes were notably lower. As displayed in figures 4 and 5, bicycle traffic for each time period has a similar pattern throughout the day, but from July 2021 to July 2022 the volumes were much lower for each day of the week. Relative to the year before the pandemic, there were significantly fewer cyclists during the weekday morning rush hour, typically from 7:00 am to 9:00 am. This may be in part due to an increase in cyclists telecommuting as opposed to commuting on the Trail. While bicycle traffic was also down during the evening rush hour, it was much closer to its pre-pandemic level, which may indicate that recreational cyclists are filling the absence of former commuters.

Figure 4

Figure 5

The rebound of pedestrian traffic at Front and Chestnut suggests an important return of economic activity to the corridor and is a promising sign for other commercial districts around the City. The route’s activity patterns have shifted throughout the week, showing a decline during weekday lunch hours and an increase in Sunday afternoon traffic.  On the Schuylkill River Trail, the decrease of bicycle volumes, especially compared to the complete rebound among pedestrians, suggests a behavioral change among cyclists. Along with the prevalence of telecommuting, the re-opening of MLK Drive to motorists, and the closure of the Chestnut Street Bridge, other factors that may be contributing to depressed bicycle volumes might include inclement weather, new COVID-19 variants and case counts, an uptick in out of town travel, or cyclists choosing other bike routes, such as the increasingly popular Delaware River Trail or the protected bike lanes on 22nd Street. For Front & Chestnut and the Schuylkill River Trail, along with many other locations through the City, future analysis is needed to determine whether the COVID-19 pandemic has permanently or just temporarily changed pedestrian and bicycle behavior in Philadelphia.

In November 2020, DVRPC and the Philadelphia’s Office of Transportation, Infrastructure, and Sustainability (oTIS) and the Streets Department completed the installation of in-street permanent bicycle counters on Spruce and Pine streets. Located near where the streets intersect 12th Street, these permanent counters fill critical data gaps that arose when planners relied on the week-long counts conducted every few years. The counters run 24/7, 365 days a year, allowing for long-term and consistent data collection and analysis.

Previous reporting has shown increases in bicycling on regional trails in the time of COVID, but new count data from these permanent counters suggests a similar finding on bicycle lanes in Center City Philadelphia.

The graph above shows total cyclists using the Schuylkill Banks portion of the Schuylkill River Trail (SRT) in red, and the total cyclists using the Spruce and Pine street bicycle lanes in blue from late November to the end of February. SRT trail volumes and peaks were higher from November to late January, at which point Spruce and Pine Street volumes overtook the SRT.

On February 15th, the SRT closed at Chestnut Street for three months of construction, which may have reduced trail use near the permanent counter. The counter is located one block south of Chestnut Street. However, it is clear that before this date, trail volumes were lower when compared to Spruce and Pine streets. This could be explained by the seasonality of recreational riders on the SRT versus the consistency of bicycle commuters and utilitarian riders using the Center City bike lanes.

To analyze bicycle commute trends during COVID, hourly data was analyzed for Spruce and Pine streets. The graph below shows the average number of cyclists per hour passing by the permanent bicycle counters from November 2020 to February 2021. Historic data from 2011 Spruce Street counts was also averaged to provide context and comparison (data for Pine Street from this time period is not available). A limitation of the week-long counts versus the permanent counts is that the Spruce Street historic data shown here only represents average cyclists for a one-week period, while the Spruce and Pine data represents averages over a three-month period. When compared to 2011 data, bike lane usage peaked earlier in the day in 2020-2021.

Permanent bicycle counters allow planners and policymakers to better understand how various factors influence cycling. The 24/7/365 nature of the counters also allows greater certainty of how many people are actually using a facility. The counters have provided valuable data regarding commuting during COVID, and the data they produce can inform policy and plans to strengthen the region’s recovery after the pandemic. For instance, hourly data could inform a plan of targeted enforcement of “no parking in the bike lane” rules to the hours when the bike lanes are used the most.

To say it’s been an eventful year is an understatement! And we can see these ups and downs translate into different patterns of pedestrian activity across the city and are reflected in our ongoing travel monitoring efforts with our partners.

The bar chart below shows pedestrian counts from March 1st to November 30th, 2020 at 3rd and Market street (in red) and 40th and Lancaster streets (in blue). Important dates and events are highlighted as they help explain some of the peaks and valleys in the chart. In Old City, the highest number of pedestrians (5,810) was counted on November 7th, the day the election was called. People were celebrating the democratic process in the very birthplace of our democracy! Levels of walking at 40th and Lancaster stayed surprisingly consistent throughout the year, which is probably reflective of the fact that it's a hub of neighborhood activity and access to transit. It is less impacted by changes in tourism and restaurant and nightlife activity, compared to Old City.

The counter at 3rd and Market is owned by Old City District. The counter at 40th and Lancaster is owned by the Philadelphia Health Department. Both agencies allow DVRPC access to their data for monitoring and analysis. Thanks to both for their sharing and collaboration!

Regional Trails are Realizing a Boom in Pedestrian Activity

A recent cycling boom nationwide has stirred much media attention. DVRPC’s previous regional analysis over the past few months has confirmed this trend. But a comparison of trail count data from March-June 2019 and 2020 shows a similar boom in pedestrian trail activity.

From March through June 2020, pedestrians have considerably increased their trail usage compared to 2019. The graph below highlights both pedestrian and bicycle increases this year. Pedestrian use is a big part of the increase in trail use this year. The Delaware River Trail at Port Richmond saw 7,000 new pedestrians in 2020, while the Wissahickon trail saw 24,000 additional pedestrians. The Chester Valley Trail gained just over 34,000 pedestrian trail users from 2019, nearly doubling pedestrian trail use. Schuylkill Banks saw an additional 50,000 pedestrian trail users.

All five automated trail counters measured an increase in pedestrian traffic analogous with the regional increase in cycling. In 2020, pedestrian and bicyclist counts surpassed the previous year’s counts on all five trails identified in the above table. The Chester Valley Trail and the Delaware River Trail at Port Richmond, saw a shift in the most common trail users with pedestrians becoming more common in 2020 than cyclists.

While we’re seeing an increase in trail use within the region, we’re also seeing a more pronounced period of use throughout the day, concentrating in the late afternoon. The graphs below offer a snapshot of hourly weekday usage averaged across March-June, 2019 and 2020 on the Port Richmond trail. In 2020, there is a peak in users between 4 and 7 PM, with softer pedestrian “spikes” at 6 AM and 12 PM than in 2019. Overall, compared to 2019, usage has increased the most dramatically in the late afternoon to evening. This information is consistent with more people working from home and using trails after the work day, for recreation, than for more traditional bicycle commute times. 

Previous analysis has shown a cycling boom, but this increase has also been mirrored by a substantial pedestrian increase on regional trails. Explore DVRPC's Permanent Bicycle and Pedestrian Count web page for more details!

Regional Trail Use Continues to be Up From Last Year

DVRPC's permanent trail count data from May 2020 shows yet another month of increased trail use in the region from 2019. DVRPC has monitored trail use during the pandemic, recently reporting on trail increases in both March and April of this year compared to last year. In fact, data shows it's been up since January 2020. The following graphic shows this uptick from 2019 to 2020 of bicycle and pedestrian trail use on four trails in the region: Chester Valley Trail, Delaware River Trail at Port Richmond, Schuylkill River Trail at Spring Mill and Wissahickon Trail.

In May, trail use was up slightly over the three-day Memorial Day weekend, despite this year's stormy Saturday. Combined data from the Schuylkill Banks, Chester Valley Trail, Washington Crossing, Port Richmond, and the Wissahickon counters show usage up during the three days by 9%. Sunday showed the biggest increase in use on the trails with usage up 52% from 2019.

As we’ve previously reported, trail use on typically less used trails is realizing significant increases. Over the Memorial Day holiday weekend for example, the Delaware River trail at Port Richmond trail saw a 132% in use over the weekend. Washington Crossing saw a 9% increase over the three-day weekend with a 68% increase on Sunday alone.

Explore DVRPC's Permanent Bicycle and Pedestrian Count web page for more count data!

Regional Trail Use Continues to be Up From Last Year

Spring 2020 trail use continues to be high. Last month, DVRPC shared the spike in trail use we saw in March using data from our permanent trail counters. April data looks the same: trail use continues to be higher on most trails compared to 2019 despite worse weather. April 2019 had roughly the same amount of rainy days (5 versus 4), but twice as many days over 60 degrees (26 versus 13). Each month had 8 weekend days, typically the highest use days.

Despite last year's better weather, two of the region's traditionally more lightly trafficked trails, the D&L at Washington Crossing and the US 202 Parkway trail, saw large percentage increases (118% and 138% respectively). The Schuylkill Banks trail, the region's most heavily trafficked trail segment, actually saw an 18% drop in use. This drop could be due in part to the weather, but the continued high use of other trails suggests users may be seeking out more lightly trafficked trails in order to maintain physical distance.

A shift in the time of day cyclists are using our trails is also occurring. As more of our region's workforce works from home rather than commuting to their workplace, usage has shifted away from the double peak "rush hours" more typical on trails used for commuting to a single "recreational" peak in the afternoon. Wissahickon trail weekday bicycle data illustrates that the April 2019 bicycle "rush hour" peaks at the beginning and end of the workday shifted in April 2020 to a single, late afternoon peak profile, more typical of recreational users.

As we celebrate Bike Month this May, continue to stay safe. Follow national guidance on physical distancing such as wearing a mask and maintaining a physical distance of at least 6' from others. Consider going on a ride in the morning instead of the afternoon, and go the extra mile by waiting to pass, signalling your approach, and maintaining a single file in some locations.
Explore DVRPC's Permanent Bicycle and Pedestrian Count web page for more details!

Read about March 2020 trail counts - 4/6/20

Project counts are bicycle and pedestrian counts that aim to tell planners, engineers and stakeholders how much bicycle and/or pedestrian activity exists in an area (intersection, corridor, etc.). If taken in the same location at different times, for instance, before or after a bike lane is striped, these counts can measure the effect on bicycling or walking.Typically DVRPC counts in locations where we, or our partners, are doing planning, engineering, or public engagement work along a corridor or site. Counts are typically conducted with pneumatic tubes or infrared devices for seven days to capture the variation across the week. For counts taken by video, a shorter time period may be required.

DVRPC's cyclical pedestrian and bicycle programs conduct counts at the same location at regular intervals, on an ongoing basis. Locations were selected to represent a range of facility and land use types across our region. DVRPC's cyclical bicycle program has about 90 locations in PA and about 60 in New Jersey. Our cyclical pedestrian program has 170 locations in Pennsylvania only. Each program started with a "saturation" year or base year where each location was counted. Subsequently, each location is counted again, on a three-year cycle. The representative sampling of locations, taken over regular intervals, allows for measuring trends across land use, facility type, and by location.

Screenline and Cordon Counts: These are five year cyclical counts taken on roads that cross the border of the DVRPC region, cross the Delaware and Schuylkill rivers, and that cross over county boundaries. Counts are also taken on roads that cross the border for the Philadelphia Center City District. These counts are used to compare and create travel trends in the region used in our model.

DVRPC also collects data from trail use counters located on Circuit Trails throughout the region. This data records trail use by direction and by mode (walking vs. bicycling) continuously throughout the course of a year. Trends in trail use can be analyzed since the inception of the program in 2008. Permanent counts also allow staff to develop seasonal adjustment factors which can be applied to project and cyclical counts to get the annual average daily bicycle or pedestrian count. The William Penn Foundation, Pennsylvania Environmental Council, and Schuylkill River Development Corporation provided generous support for the acquisition and installation of the counters.

Average Daily Bicycle (AADB) or Pedestrian (AADP) volumes are calculated by summing each factored daily count and then dividing by the number of full days in the count (usually 7 days). To get the Annual Average Daily Bicyclists or Pedestrians, seasonal factors are applied to the raw count data available.

DVRPC's permanent counters provide yearlong data through documented weather conditions, seasonal and weekday variations. By analyzing this data for recurring changes associated with the calendar day over multiple years, DVRPC can measure the percentage amount that on average, a count is above or below normal. These seasonal adjustment factors are calculated for each day of the week within each month and range from .763 to 1.4. Daily totals from the observed count are multiplied by a seasonal adjustment factor and an equipment factor (provided by the manufacturer) and result in the Annual Average Daily Bicycles (AADB). An equipment factor attempts to correct for the average amount the equipment undercounts or misses users.

For example: If a bicycle count on a Monday in March is 500 with a seasonal factor of .95 taken on ECO Counter equipment that has a 1.1 equipment factor then the corrected total or AADB, is 522 (500 * .95 * 1.1).

Annual Average Daily Pedestrians (AADP) volumes are calculated similar to bicyclists but since we don't yet have the recurring data for pedestrians associated with the calendar day over multiple years, we use factors from vehicular traffic on roadways. These vehicular traffic factors take into account route type, month of the year, and day of the week. Once the raw pedestrian count is multiplied by the roadway factor and equipment factor, all factored days are summed and averaged for the Average Daily Pedestrian count.