This visualization shows different assets (Jobs, Transit Stations, Roads, etc.) that are impacted by flooding. The data is derived from GIS intersection analyses. Jobs, Population and Household data were only available at a zonal level; therefore, impacts were determined using a proportional split procedure where the proportion of inundated area in a zone was used to determine the proportion of impacted assets in that zone. All other data was available as point data that had specific X & Y coordinates. These data sources were directly intersected with the different water layers to determine if they were potentially impacted.

Use:

• The buttons trigger different data sources and corresponding graphs. The main data sources are Demographics, Transit and Auto.
• The Water button toggles on and off the water layers.
• The slider bar transitions the data for the different inundation levels.
• There are two tabs to the right of the main map: Charts & Specifics
  • “Charts” contains small charts showing the changing impact over the different inundation levels
  • “Specifics” contains a grouped bar chart highlighting a specific detail of the data related to either Demographics, Highway, or Transit
• All data can be normalized by pressing the Normalize button.

Main Findings

• By examining the grouped chart under the specifics tab in the Demographics data we see that the percentage impact of inundation on service jobs is very high. Service jobs are arguably the most important sector to the economy and are growing faster than retail and basic jobs This highlights that flooding has the potential to cause major economic impacts on the city of Boston just from an employment perspective.
• In terms of Transit, the Red Line and Blue Line see major impacts under inundation levels four feet and higher. We can expect many transit users will be unable to reach a destination if inundation has occurred in the city.

This visualization articulates the difference between ridership on transit routes at no inundation versus four feet of inundation. Inundation causes specific routes to be heavily impacted or inoperable in certain regions. The model allows portions of a route to operate even if other portions are inundated. The four-foot inundation level is presented because there are major impacts, though much of the region is still free of inundation. I do not present all six inundation levels because of data size constraints.

This visualization allows us to examine possible shifts in ridership due to inundation impacts. A possible question that can be examined with this visualization is "If the Red Line is inundated what other routes might persons use to reach their destination." Such answers could be used to preemptively bolster service on specific bus routes or rail lines.

Use

The Table

The table contains data on:

• Each transit route, and its specific mode, present in the model system
• Ridership on these routes given no inundation (Base Ridership)
• Ridership on these routes given 4ft inundation (4ft Ridership)
• Difference in Ridership (4ft Ridership - Base Ridership)
• Clicking a route in the table will display the boardings and alightings at stops along that route - as well as nearby ( 1/2 Mile ) stops that had either Ridership Increases or Ridership Decreases.

Ridership Increase

When the “ridership increase” button and a route are selected from the table, the only stops visualized will be those stops along the selected route. Stops within 1/2 mile of the selected route that had increases in ridership will also be highlighted. This allows a user to search for routes that possibly gained ridership because of inundation on another route. (See Red Line).

Ridership Decrease

When the “ridership decrease” button and a route are selected from the table, the only stops visualized will be the stops on the selected route, as well as those stops within 1/2 mile of the selected route that had decreases in ridership. This allows a user to search for routes that may have gained riders in response to other nearby routes being inundated. (See CT2).

Main Findings

• A unique opportunity to see potential shifts in route choice given inundation in the area and on the network is provided.
• Fixed rail lines see major decreases in ridership that are not accounted for in increases in bus ridership.
• The transference of riders from the Red Line to bus routes CT2 and 85 is illuminated. This increase in ridership is likely Red Line users who can no longer reach their destination on that route.

Accessibility metrics attempt to quantify the ability to reach assets (in this case Jobs) from a zone. The simplest accessibility measure is a cutoff measure which communicates the number of jobs a person in a given zone can reach within a certain amount of time. These are also called Isochrones. I have normalized each zone’s accessibility by dividing the number of jobs within each zone accessible within an hour by the total number of jobs in the model region.

There are many other accessibility metrics that vary in complexity, I have included a measure that incorporates an impedance function, or a Gamma Measure. This accessibility more heavily weighs jobs that take shorter time to access than those that take a longer time. There is no time cutoff in this measure, though at high travel times the contribution of a zone’s jobs to the total sum of weighted jobs is close to zero. The gamma measure output values can only be interpreted relative to other gamma function values.

Use

The slider will display the accessibility changes over the different inundation levels. One can view the accessibility for Walk, Transit and Auto. The scale for each mode is fixed to facilitate comparison over inundation levels.

The Accessibility Histogram is "brushable" and will show only those zones with corresponding values in the map.

Main Findings

• Auto always has the highest accessibility and is much less impacted by inundation because of the dense and redundant road network in the region.
• Transit and Walk accessibility to jobs is always drastically smaller than Auto accessibility and is concentrated region's inner core.
• Transit accessibility to jobs is drastically impacted by inundation.

Accessibility Calculation

$$ \mathbf{Jobs\ Gamma \ Accessibility }$$

$$Accessibility_{i} \sum_{j \in Z} \frac{Jobs_i}{\sum_{j \in Z} Jobs_j} * TT_{ij}^{-.503} * e^{-0.078 * TT_{ij}} $$

$$ \text{Where $Z$ = All Zones}$$

$$ \mathbf{ Jobs\ Cutoff\ Accessibility}$$

$$Accessibility_{i} = \frac{\sum_{j \in Z} Jobs_j*TT_{ij}}{\sum_{j \in Z} Jobs_j} $$

$$ TT_{ij} = \begin{cases} 1, & \text{if $ TT_{ij}$ is $\leq$ 60 Minutes} \\[2ex] 0, & \text{if $ TT_{ij}$ is $>$ 60 Minutes} \end{cases}$$

$$ \text{Where $TT_{IJ} = $ Travel Time Between Zone $I$ and Zone $J$ }$$

$$ \text{Where $Z$ = All Zones}$$

Many trips can no longer be completed given inundation of the road and transit networks. This visualization shows those origin and destination pairs that lost the most trips. I used a cutoff value of 50 trips to prune the dataset down. The direction of flow is collapsed, as well, to reduce the size of the data. This means that trips lost between zones O & D are grouped with trips lost between zones D & O. If the total lost trips between a given Origin and Destination was less than 50 then a line is not drawn. The thickness of the lines indicates the number of lost trips. Furthermore, they are redundantly encoded by color to make them visually clear. This type of visualization is commonly called a "Desire Line" map and is typically generated in proprietary transport modeling software packages.

Use

Click the Buttons to change between Transit, Auto, Total.

Total is just the sum of Transit and Auto.

The Lost Trips Histogram is "brushable" and linked to the map. When brushed only those brushed values will appear in the map.

The Total Percent of Lost Trips is included for reference. About 16.4 Million total number of trips occur in one day in the model region.

Main Findings:

• The greatest number of lost auto trips occur north of the inner core. Though the total number summed by the inner core is likely larger, it is key to note that because the trips are split up into many different zones and zones are smaller in the inner core the values can appear smaller.
• Transit Lost trips are far fewer but very concentrated in the inner core.
• Lost trips occur where inundation occurs (reference inundated assets map)

This visualization shows the modeled highway volumes and queues of vehicles across the six different inundation levels.

Volumes

Modeled auto on volumes each link. The unit of measurement is vehicles per link.

Queues

Queuing vehicles are those vehicles that can pass through a given link because of congestion on the highway network. Areas of high queues would be areas that are heavily congested (i.e. standstill traffic).

Main Findings:

• As inundation increases volumes move to outer ring roads (MA Route 128 & HWY 95 & HWY 93)
• As inundation increases queues move toward these outer ring roads as well, though there are other links that become heavily congested in the inner core area. These links are likely congested because other alternative links are inundated.
• At six feet of inundation there is much less congestion in the network because there are far fewer vehicles in the network overall