Most maglev customers would start and end their trips near a maglev station, limiting the project’s utility to the region. The maglev’s draft environmental impact statement leaves it to the reader to piece this story together.
This is the second in a series of articles about the proposed Baltimore-Washington maglev, reconsidered in light of the project’s draft environmental impact statement that was published in January 2021. The first article in this series examined whether the maglev would significantly reduce regional road congestion. This second article examines whether the maglev would serve the whole Baltimore-Washington region or just relatively small areas near the three maglev stations.
To explore this question, the article estimates how far from the maglev stations most maglev customers would start and end their trips based on information in the draft environmental impact statement and other documents. One approach to this question is to calculate how much time the maglev would save people relative to the time that would otherwise be spent driving directly to the destination. It is disappointing that the draft impact statement itself lacks a map to show where most maglev customers would start and end their trips.
A magnetic-levitation rail line has been proposed to connect Baltimore and Washington. To evaluate this proposal, it would help to have a sense of where most of this rail line’s customers would start and end their trips.
It is important to know where most “maglev” customers would start and end their trips because much of the economic benefit from operating the maglev may concentrate in the same area. The maglev’s draft environmental impact statement (DEIS) is vague about how geographically concentrated would be the maglev ridership and economic benefits. Would they be concentrated tightly near maglev stations or spread out more evenly over a large region?
The DEIS quantifies the maglev’s economic impact over vast area, the Washington-Baltimore Combined Statistical Area. In contrast, the DEIS calculates its forecast for the maglev’s ridership over a smaller, but still large, area that is defined by a 25-mile radius around each maglev station. These areas are shown in Figure 1 of the present article.
As explored in the present article, even a 25-mile radius seems like an overestimate of the maglev’s reach. Instead, most maglev customers would probably start and end their trips within a small subset of the 25-mile-radius area around each maglev station. The present article suggests that most counties in the Washington-Baltimore Combined Statistical Area would have few if any maglev customers starting or ending their trips there.
Elected officials and the public would like to know which counties and cities would benefit from the maglev and which would be harmed by its construction and operation. For example, the DEIS estimates that 390 to 440 jobs would be created directly or indirectly as a result of operating the maglev, but the DEIS is silent on the question of where these jobs would be located. The maglev’s many negative impacts are quantified in various sections and appendices of the DEIS.
It would have been helpful if the DEIS had plotted contours on a map or used some other means to visualize where most maglev customers would start and end their trips. The public is unable to find this information in official sources, such as the studies, memos, and data requests that are the source of the DEIS’s maglev ridership forecast. These documents are hidden from public view. Their existence is known only from footnotes in the DEIS. Fortunately, enough information is published in the DEIS to guide the analysis in the present article.
In January, 2021, the Federal Railroad Administration published the draft environmental impact statement (DEIS) for the proposed Baltimore-Washington maglev. The DEIS states that using the maglev would save a traveler 8 to 27 minutes relative to the time that the traveler would otherwise spend driving directly to his or her destination.
Because maglev tickets would be so expensive, it is plausible that people would ride the maglev only if it saved them at least 8 to 27 minutes. The DEIS states a ticket price of $40 to $80 per person, one way. The cost of driving between Baltimore and Washington is approximately $7 per car, one way, based on the average trip length stated in the DEIS and the AAA estimate for the cost of fuel and maintenance for a typical car. As a result, the maglev-vs.-car price difference is $33 to $73, one way, with one person in the car, and much more than $33 to $73 with multiple people in the car, such as on a date or family outing.
Travel-time saved and travel cost are factors that transportation planners consider when forecasting the ridership for a transportation proposal. The DEIS states that these factors were included in the model that forecast the ridership for the proposed Baltimore-Washington maglev.
The present article identifies the maglev ridership area by exploring where the maglev would save a customer approximately 8 to 27 minutes relative to the amount of time that the customer would otherwise have spent driving directly to the destination.
To estimate travel-time saved, first pick a trip origin and destination with one of the points in the Washington area and the other point in the Baltimore area. Calculate the time to drive between these two points. From that time, subtract the time that it would take to travel between the same two points using the maglev, which in simplest form would mean driving to a maglev station, riding the maglev, and riding a car to the destination. Various online applications can provide car travel time between any two points, and time spent on the maglev itself can be estimated from information in the DEIS and other documents. The details of this calculation are described in Appendices 1 and 2 of the present article, which are available in the PDF version of this post.
The present article estimates travel-time saving for a scatter of trip origin-and-destination pairs in order to build up knowledge of where on the map the maglev travel-time savings would be in the 8-to-27-minute range that the DEIS provides. The computations are made slightly more complicated because two stations are proposed at the Baltimore end of the trip, one at Camden Yards and one at Baltimore/Washington International (BWI) airport. The solution is to calculate travel-time saved for one or the other Baltimore maglev station, and then use whichever value is greater.
One simplification employed in the present article is to assume that travel to and from the maglev stations occurs by car, without modeling the option of subway travel to and from the Washington maglev station. Supporting this simplification, the analysis in Appendix 3 of the present article finds that, in almost every case, a subway ride would not save time over driving to the downtown Washington maglev station. The existence of the Washington subway has little impact on the geographic extent of the maglev ridership area.
The calculation method used here is kept simple because the goal is merely to determine whether the maglev ridership area would fill the entire 25-mile-radius area that is studied in the DEIS or if the ridership area would be much smaller than that.
The three sections below identify jurisdictions in which most maglev travelers would start and end their trips during rush hour or in light traffic. Also identified are jurisdictions with little area or no portion of them served by the maglev regardless of the road-congestion level. One finding is that the proposed Baltimore-Washington maglev would have an easier time competing against car travel during rush hour than when road traffic is light. In other words, the maglev ridership area is larger during rush hour than when road traffic is light.
The Maglev Ridership Area during Rush Hour
How far one can travel from a maglev station and still save 8 to 27 minutes depends on how close the other end of the trip is to the other maglev station. Figures 2 and 3 show two possible realizations of the maglev ridership area during rush hour. One option emphases access to the Washington area and the other option emphases access to the Baltimore area.
Figure 2 emphases locations at the Washington end of a rush-hour trip while still reaching an appreciable number of locations in the Baltimore area. Optimized in this way, the maglev ridership area would include about half of the District of Columbia; most of the City of Alexandria, Arlington County, and City of Baltimore; and less than half of the Baltimore County suburbs.
In contrast, Figure 3 shows the maglev ridership area optimized in the opposite way. Figure 3 emphasizes locations at the Baltimore end of the trip. In this case, a portion of eastern Carroll County and northern Anne Arundel County can be reached. This portion of Carroll County is sparsely populated and this portion of northern Anne Arundel County contains Glen Bernie and Pasadena. Few people would make use of the maglev in this scenario because only a small portion of the District of Columbia can be reached at the other end of the trip. The following portions of the District of Columbia cannot be reached: the Capitol building, Capitol Hill, most residential areas in the District of Columbia, and the federal offices just south of the National Mall.
The DEIS forecasts that approximately 15% of maglev travelers would be airline passengers headed to or from BWI airport, and the present article neither confirms nor questions that forecast.
The present article does, however, suggest that the BWI maglev station would have limited utility for points other than the airport’s main terminal. Figure 2 shows a small maglev ridership area that is located just to the north and east of BWI. From the rest of the business parks and residential areas within a few miles of BWI, one can easily reach Interstates 95 and 295, which are direct routes to Washington. Starting from these locations, one can reach the BWI main terminal from only one direction (the west) and the airport’s main road loop can be slow due to congestion. In this way, the existing road network would geographically isolate a maglev station adjacent to the BWI main terminal.
To summarize the rush-hour results, the maglev would save the traveler approximately 8 to 27 minutes over an area that is much smaller than the area of the DEIS-supplied 25-mile radius about the maglev stations.
Careful examination of Figures 2 and 3 reveals that the maglev ridership area is bunched to the side of the maglev station that is furthest from the other maglev station. In other words, the maglev ridership area is mostly south and west of downtown Washington and north and east of downtown Baltimore. This makes sense because you aren’t going to save much time using the maglev if your trip starts and ends between the two cities. In this case, traveling to and from the maglev station would take you far out of your way.
Another thing evident in Figures 2 and 3 is that the maglev ridership area is larger at the Baltimore end of the trip than at the Washington end of the trip. This asymmetry is due to the fact that the proposed Washington maglev station at Mount Vernon Square would be stuck in the middle of an area with especially slow rush-hour traffic and many traffic lights. In contrast, the maglev station proposed for Baltimore’s Camden Yards would be a short detour from routes that would take drivers initially south and west toward downtown Baltimore along Route 83 and Interstate 95 and subsequently south toward Washington.
The present article excludes Hartford County from the maglev ridership area at the Baltimore end of the trip. Slightly less than half of Hartford County is within the DEIS’s 25-mile radius from the maglev station proposed at Baltimore’s Camden Yards. The lack of an existing market for the maglev in Hartford county is indicated by US Census data. Census data shows that almost no Hartford County residents commute to jobs in Washington and almost no Washington residents commute to jobs in Hartford County. Following the same sort of logic, the DEIS states that it shrunk or expanded its 25-mile-radius area, as necessary, to reflect existing travel patterns.
The Maglev Ridership Area when Road Traffic is Light
In light traffic, the maglev would save travelers 8 to 27 minutes over an even smaller area than it would during rush hour. Two factors contribute to the maglev’s limited utility when road traffic is light. First, directly driving to the destination is much faster in light traffic than in rush-hour traffic. Second, maglev trains would be less frequent outside of rush hour, and therefore one would wait longer for the next maglev train. These two factors influence the size of the maglev ridership area shown in Figure 4.
When road traffic is light, the utility of the maglev is limited in several ways. First, only a small portion of downtown Washington and downtown Baltimore would be included in the ridership area, as shown in Figure 4.
Second, Appendix 2 of the present study suggests that the maglev would save travelers only 10.5 to 17.5 minutes of travel time when road traffic is light, i.e., the lower half of the target range of 8 to 27 minutes. Such limited travel-time savings suggests that, outside of rush hour, only wealthier maglev riders would find the maglev travel-time savings sufficient to justify the $40 to $80 maglev ticket price.
Last, the ridership area depicted in Figure 4 applies only to various non-rush-hour times during which there are maglev departures at least every 15 minutes. In contrast, maglev departures that are 30 minutes apart may occur during off-peak weekend hours, and at these time, the maglev ridership area would essentially disappear.
Jurisdictions Not Served by the Maglev
The present article finds that many counties are outside of the area served by the maglev but are included in the DEIS study area. Based on the travel-time analysis in the present article, elected officials and members of the public should read with skepticism any claim that the Baltimore-Washington region, as a whole, would benefit from the maglev rather than a few small areas near a maglev station.
Even during rush hour, few if any maglev customers would start or end their trip in the majority of the counties within the jurisdiction of the Metropolitan Washington Council of Governments or the Baltimore Metropolitan Council. These counties are outlined in white in Figure 1 of the present article. In addition, few maglev customers would start or end their trips in the majority of the counties in the Washington-Baltimore Combined Statistical Area, which is also shown in Figure 1. In fact, few if any maglev customers would even pass through these counties on their way to or from a maglev station.
The present analysis compares travel time when a trip is made using the proposed maglev between Baltimore and Washington or is made entirely by car. The results of the analysis are maps of the maglev ridership area, the area near maglev stations where most maglev customers would start or end their trip.
In the present article, the maglev ridership area is modeled as the area where the maglev would save a traveler approximately 8 to 27 minutes compared to the time that the traveler would otherwise spend driving directly to his or her destination. The maglev’s draft environmental impact statement (DEIS) asserts that a maglev customer would save this much time. More importantly, travel-time savings at least this great are a plausible prerequisite for people who travel between Baltimore and Washington to find the maglev to be an attractive option in light of the maglev’s $40-to-$80 ticket price per person, one way.
During rush hour, the present article finds that the maglev would save travelers about 8 to 27 minutes on trips that start and end in at least half of the area of each of these jurisdictions: the District of Columbia, the City of Alexandria, Arlington County, Baltimore County suburbs, and the City of Baltimore. Even during rush hour, few if any maglev customers would start or end their trips in the majority of the counties within the jurisdiction of the Metropolitan Washington Council of Governments or the Baltimore Metropolitan Council.
When road traffic is light, the maglev ridership area would be even smaller. It would include, at most, only a portion of downtown Washington and downtown Baltimore. The reason for the maglev’s limited utility when road traffic is light is that there would be fewer maglev trains per hour than during rush hour and car travel would be much faster than during rush hour.
During both rush hour and periods of light road traffic, most maglev riders would start and end their trips in a small portion of the 25-mile-radius area about each maglev station, the 25-mile radius that the DEIS used in its ridership forecast. The present article comes to this conclusion, and the DEIS neither confirms nor denies it. Elected officials and the public should investigate for themselves whether the maglev’s forecasted economic benefits are realistic given that the DEIS forecasted these benefits over such a large area and the present article finds that maglev ridership would be concentrated over a much smaller area.
AAA, 2020 Dec 14: Your Driving Cost: 2020. 8 pp, https://newsroom.aaa.com/wp-content/uploads/2020/12/2020-Your-Driving-Costs-Brochure-Interactive-FINAL-12-9-20.pdf.
Baltimore Metropolitan Council (BMC), 2020: Maximize2045: A Performance-Based Transportation Plan. https://www.baltometro.org/transportation/plans/long-range-transportation-plan/maximize2045.
Department of Transportation (DOT), 1997: Transfer Penalties in Urban Choice Modeling. published by the Travel Model Improvement Program, DOT-T-97-18, 51 pp., https://babel.hathitrust.org/cgi/pt?id=ien.35556028271757.
Department of Transportation (DOT), 2011: High-speed/Intercity Passenger Rail (HSIPR) Best Practices: Ridership and Revenue Forecasting. 89 pp., https://www.oig.dot.gov/sites/default/files/files/OIG-HSR-Best-Practice-Ridership-and-Revenue-Report.pdf.
Encyclopedia Britannica, 2020: Escalator. https://www.britannica.com/technology/escalator.
Federal Railroad Administration, 2021 January: Baltimore-Washington Superconducting MAGLEV Project Draft Environmental Impact Statement and Draft Section 4(f) Evaluation. 654 pp., https://bwmaglev.info/index.php/project-documents/deis.
Guo, Z., and N. H. M. Wilson, 2004: Assessment of the Transfer Penalty for Transit Trips. J. Transportation Research Board, No. 1872, 10–18, https://journals.sagepub.com/doi/pdf/10.3141/1872-02.
Kelley, O., 2021 Feb. 10: Maglev would do little to reduce road congestion, says Federal Railroad Administration. blog post, Greenbelt Online, https://www.greenbeltonline.org.
Kelley, O., 2021 March 25: The proposed Baltimore-Washington maglev would serve a small geographic area. PDF file, 26 pp., https://www.greenbeltonline.org.
Metropolitan Washington Council of Governments (MWCOG), 2010: Region Forward Vision. 72 pp., https://www.mwcog.org/regionforward/.
Microsoft, 2018: Bing Maps Route API, https://docs.microsoft.com/en-us/bingmaps/rest-services/routes/.
Ortuzar, J. de D., and L. G. Willumsen, 2011: Modeling Transportation. 4th ed., Wiley, 586 pp.
Roy, S., 2017 Dec. 07: How many people commute between Baltimore and DC? Greater Greater Washington blog, https://ggwash.org/view/65822.
Titus, J., 2015 Feb. 02: Governor Hogan thinks only 10% of Marylanders use transit. Actually, 25% or more do. Greater Greater Washington blog, https://ggwash.org/view/36978.
US Census Bureau, 2015: Table 4, Residence MCD/County to Workplace MCD/County Commuting Flows for the US and Puerto Rico Sorted by Workplace Geography. 2011–2015 5-Year American Community Survey (ACS) Commuting Flows, https://www.census.gov/data/tables/2015/demo/metro-micro/commuting-flows-2015.html.
US Census Bureau, 2017: American Community Survey Information Guide. ACS-331(C)(2017), 18 pp., https://www.census.gov/content/dam/Census/programs-surveys/acs/about/ACS_Information_Guide.pdf.
Willen, C., K. Lehmann, and K. S. Sunnerhagen, 2013: Walking Speed Indoors and Outdoors in Healthy Persons and in Persons with Late Effects of Polio. J. Neurology Research, 3, 62–67, doi: https://dx.doi.org/10.4021/jnrl187w.
Willumsen, L., 2014: Better Traffic and Revenue Forecasting. Maida Vale Press, 258 pp.
 CSA: Chapter 4.6, pg. 4.6-1; 25-mile radius: Appendix D2, pg. C-106.
 390-440 jobs: Chapter 4.6, pg. 4.6-8. Negative impacts would occur to the following areas of kinds of resources: historical sites (Chapter 4.8); scenic resources (Chap. 4.9); recreational facilities (Chap. 4.7); environmental justice (Chap. 4.5); quality-of-life (Chap. 4.4); hazardous waste sites (Chap. 4.15); forests, forest-interior species, and habitats of rare, threatened, and endangered species (Chap. 4.12); wetlands (Chap. 4.11); economic harm during construction (Appendix D4, pg. D-18 to D-30); and lost revenue for Amtrak and MARC commuter trains (Appendix D4, Table D4-47, pg. D-54).
 DEIS: FRA 2021; 8-27 minutes: Appendix D4, pg. C-6.
 The DEIS considered and rejected a maglev ticket price as low as $27 and chose instead to base its ridership forecast on a $40–$80 ticket price: Appendix D2, pg. D-107, D-108; $7.08 cost of making a typical trip between Baltimore and Washington by car based on a 39.6-mile trip length (Appendix D4, Table D4-59, pg. E-82) and a $0.1787-per-mile cost for medium sedan (AAA 2020).
 Willumsen 2014, Chapter 5; Ortuzar and Willumsen 2011, Section 15.4; Ridership forecast model: Appendix D2, pg. B-104 to E-110.
 Only 14.5% of maglev trips would be downtown-to-airport, with the remaining 85.4% downtown-to-downtown: Appendix D4, Table D4-25, pg. D-42.
 25-mile radius: Appendix D2, pg. C-106; US Census Bureau (2015).
 Map of counties in MWCOG (2010, pg. 5) and BMC (2020, pg. 6).
About the Author: Owen Kelley has a science background, and in his free time, he enjoys exploring and writing about the forests around Greenbelt. In recent years, he has written several articles about the proposed Baltimore Washington maglev.
Disclaimer: Kelley is writing in his capacity as a individual citizen examining a non-partisan issue of interest to the public. If errors are suspected, please contact him at email@example.com.