READ-AHEAD MATERIALS FOR RESIDENTIAL PARKING WORKING GROUP MEETING FOUR, PART ONE In preparation for Residential Parking Working Group Meeting Four, staff have assembled the following categories of information: Findings from surveys of parking utilization and trip-making at multi-family residential Site Plan buildings in Arlington. Data on vehicle access among Arlington households and the modes of transportation that Arlington residents choose for their trips from various sources. Due to strong interest from Working Group members on the relationship between household income and vehicle access (as well as mode choice) we have assembled data on these sub-topics as well. Research on the relationship between parking demand and other factors. Note that we will post additional data the Documents page on Monday, October 28 th, 2016. We ask you to read this second set of data before Meeting Four. Additional background materials for extra reading can be found in the Additional Readings section of Meeting Four on the Documents page of the project web site.
Highlights from the 2013 and 2016 Arlington Residential Building Studies 2013 Residential Building Aggregate Study (Published Final) Results Related to Vehicle Ownership and Use This analysis was based on resident surveys, vehicle trip counts, and parking occupancy calculations. Key findings include: Vehicle ownership increased with average household income. Condominium owners owned more vehicles per adult than apartment residents. There was an inverse relationship between vehicle ownership and transit access/proximity. Ownership rates were lower in more walkable areas than in car dependent areas, but were about the same if the area was somewhat, very, or extremely walkable. Vehicle ownership was related to the cost of residential parking particularly at a cost of $95 or more per month, vehicle ownership dropped. Few parking garages approached full occupancy. The average maximum parking occupancy for all study buildings was 80%. The average minimum parking occupancy was 38% within the Metrorail corridors and 20% outside the Metrorail corridors. Parking occupancy and vehicle use seemed unrelated to the spaces per resident provided. Overall parking occupancy within Metrorail corridors was similar for all weekdays. Weekend occupancy was higher. Sunday evening occupancy was similar to the occupancy on weekday evenings. Across all study respondents, the average number of vehicles was 0.84 per adult resident. Condominium residents had a higher vehicle ownership rate (0.88 vehicles per adult) than apartment residents (0.79 vehicles per adult). Vehicle Availability and Annual Household Income NOTE: TREND LINES ARE ILLUSTRATIVE ONLY, NOT A REGRESSION LINE OF BEST FIT.
Garage Occupancy and the Building s Location Relative to Metro Service 100% 90% 80% 70% 60% 50% 40% 30% Max Parking Occupancy Min Parking Occupancy 20% 10% 0% < 0.2 miles from Metrorail > 0.2 miles from Metrorail; in Metro Corridor Outside Metro Corridor When grouping by distance from Metrorail, being outside the Metrorail corridors did not seem to affect the maximum parking used (i.e., when all cars are in for the night), but did seem to impact the minimum parking occupancy during the day (i.e., residents outside the Metrorail corridors appeared to use their cars more during the day, probably for commuting, whereas in the corridors more cars sat all day in the garages). Other Observations 1. Conversions of buildings from condo to apartment or vice versa could impact the right amount of parking for a building, independent other factors. 2. This building sample was not large enough to allow for disaggregation to individual corridors. Study description Transportation performance monitoring studies were conducted at 16 residential sites between 2010 and 2012. Eleven sites were located within the two Metrorail corridors, i.e. Jefferson-Davis or yellow/blue line and Rosslyn-Ballston or orange line. Of the five remaining sites, one site was located near the East Falls Church Metrorail but was aggregated with sites outside the Metrorail corridors since many neighborhood and travel variables were similar to these sites. One site was located in Shirlington and another along Columbia Pike, two planning areas with rich bus service but no Metrorail service. Two sites were located on either side of I-395 close to the Glebe Road exit. Neighborhood data for each site included the census blocks within Arlington whose centroid was less than a quarter mile from the study site. The 16 buildings included seven apartment buildings, one extended-stay hotel, and eight condominium buildings. The buildings varied in population density, and some had retail on the ground floor. This sample represented about: 3,700 occupied dwelling units at an average occupancy of 96 percent 4,840 total parking spaces, all types 1.04 1.55 residential parking spaces per unit (not including visitor/retail spaces) Over 38,000 vehicle trips 1,450 resident survey responses (25 percent response rate)
Data collection PARKING AND TRIP GENERATION Vehicle trips were counted by tube (or hose) counts for 24 hours/day for seven consecutive days for each entrance/exit of parking facilities, i.e. garages or surface lots. Tube counts are a widely used method for automatic trip data collection. Counts are conducted by placing a rubber tube or hose across the travel lanes, and recording the pressure changes caused by wheels of vehicles crossing the tubes as axle movements. Trips were aggregated into 15-minute intervals. Parking occupancy was calculated for the seven-day survey period based on a one-time manual count during the week. The counts were used to identify key variables: Peak hour time of day (AM and PM) Peak hour trips generated Daily total trips generated Parking occupancy by time of day The counts were compared with ITE codes 221 (low-rise apt); 222 (high-rise apt); 232 (high rise condo/townhouse); 310 (hotel). RESIDENT SURVEY Resident participation in the survey was voluntary and surveys were conducted both online and on survey forms disseminated by the survey team at on-site events or through the property manager. The property manager was also asked to send notifications and reminders over e-mail for a period of two to four weeks or until a response rate of at least 20 percent was reached. The surveys were used to assess the following key variables: Weekly commute mode split, commute distance, and other commute characteristics Mode share of non-work trips Vehicle ownership
2016 Residential Building Aggregate Study (Updated Sample, New Analysis Draft Results Only) Findings Related to Parking Supply and Occupancy Parking Supply and Occupancy at 37 Arlington Multi-Family Residential Buildings (Sorted by Maximum Parking Occupancy) Building Name Total Parking Spaces Parking Ratio per Unit Parking Ratio per Resident Unbundled Parking Max Parking Occupancy Min Parking Occupancy The Macedonian 44 1.2 1.00 Yes 45% -5% The Halstead 460 1.1 0.81 Yes 48% 9% Crescent Falls Church Apartments 398 1.6 0.78 No 52% 8% Garfield Park 282 1.2 0.80 Yes 53% 19% Siena Park 410 54% 26% The Phoenix at Clarendon Metro 501 1.2 0.79 Yes 56% 29% Liberty Center Residences 312 1.3 1.12 Yes 66% 40% Millennium at Metropolitan Park 349 1.1 0.74 Yes 70% 30% Odyssey Condominium 340 1.2 0.81 No 74% 37% Gramercy at Metropolitan Park 460 1.09 0.72 Yes 74% 42% Clarendon Center 458 1 0.74 Yes 75% 40% 1800 Wilson Blvd 192 1.3 1.10 Yes 75% 32% Grove At Arlington 297 1.5 1.12 No 77% 25% V Point 120 1 0.63 Yes 78% 32% Monroe Condominium 132 1.4 0.77 No 79% 42% The Shelton 108 1.1 0.71 Yes 80% 18% Penrose Square 428 1.4 1.10 Yes 82% 30% Pershing Apartments 271 1.4 0.86 Yes 83% 14% Dolley Madison Towers 478 1.3 0.66 Yes 83% 5% Quincy Plaza 615 1.2 0.82 Yes 84% 43% The Continental Condominium 470 1.1 0.78 No 84% 37% Turnberry Towers 379 1 0.84 Yes 84% 36% ZosoFlats 212 1.3 0.84 Yes 85% 32% 55 Hundred Apartments 315 1.2 0.72 Yes 90% 31% ARC 3409 (Joule) 91.56 1 0.88 No 90% 23% Buchannan Gardens Apartments 92 0.8 0.42 No 92% 35% Parc Rosslyn 255 1.1 0.75 Yes 96% 45% Crystal City Lofts 234 1.3 1.19 Yes 100% 49% The Palatine 303 1.2 0.76 Yes 101% 25% The Jordan 77 0.8 0.41 Yes 101% 40% Sedona Slate 422 0.9 0.55 Yes 110% 54% IoPiazza 339 1.3 0.63 Yes Westlee Condominium 211 No Liberty Tower Apartments 272 1.1 0.71 Yes Lofts 590 221 1 0.72 Yes Rows in bold are buildings that were included in 2013 report. Blank cells represent missing or not yet finalized data.
Other Findings 92% of the residents that have a car park it in the residential building parking and 83% of those that have a second car park it there too. 42% of the residents report a parking fee less than $50/month and 76% less than $100/month. Half of the residents that park at work do it for free; only 32% of non-sov users in commute trips that occasionally drive to work do it for free. In consequence, 41% rate the availability to park at work as very good, but only 27% non-sov users do. Study Description and Data Collection The final dataset used to conduct the analysis contains over 2,900 individual responses from residents of 37 buildings. One building (condos) was combined with its neighbor (townhomes) due to the design of the site and the response rate. Hence, reporting at the building level includes description of 36 sites. Data from both primary and secondary sources are included in this analysis. The three primary data collection efforts consisted of: 1. An interview with the property managers to obtain building level data such as the number of units and building occupancy, availability of parking, and types of TDM services offered. Historically the property manager interview included questions about familiarity with Arlington County services as well. 2. A resident travel behavior survey was administered to all the residents of each building included in the study. 3. On-site data collection consisted of intercept surveys and entrance/exit counts, and parking counts at each building site. The parking protocol was to have a full count of garage entrances and exits 24 hours a day for seven days. Most counts were made using pneumatic tubes, in cases where the garage entrances and/or exits presented a difficult geometry, video monitors were used instead. The intercept survey was administered to persons entering and exiting the sites and asking them only the transport mode they had used to reach the building or the mode they were about to use upon egress. These data were supplemented with secondary source data such as Walkscores, Transitscores, county travel data and US Census commute data.
Vehicle Access and Mode Choice in Arlington Staff have created a series of tables and charts to describe vehicle access in mode choice in Arlington over time. These data come from a variety of sources, including the Census Bureau, the Metropolitan Washington Council of Governments, and Arlington s own original data-collection efforts as conducted through Arlington County Commuter Services. How many vehicles do Arlington households have? US Census Bureau data is one source of information that allows us to see vehicle access patterns given current conditions. Grouping households by where they are located in the County allows us to see differences between the two Metro Corridors and the County as a whole. As seen in this chart, zero and one-car households are more common in the Jefferson Davis Corridor than the Rosslyn-Ballston Corridor, though households in both corridors are less likely to have cars than all Arlington households. DATA SOURCE: AMERICAN COMMUNITY SURVEY 2010-2014. NOTE THAT THE CONCEPT OF OWNERSHIP OR ACCESS IS TECHNICALLY ASKED AS AVAILABILITY.
How does vehicle access vary with household income? The Census Bureau does not publish as part of its usual releases data that describes vehicle access grouped by household income. A special program (known as the Census Transportation Planning Package) does publish this data, though the most recent release covers data collected between 2006 and 2010. Here we present that data for Arlington County as a whole; it is not possible to obtain US Census data that describes vehicle access by household income in small enough geographies so that someone can look at how the data changes depending on distance to Metro. Households with higher incomes have more vehicles than those with lower incomes. Vehicle Availability by Annual Household Income, All of Arlington 100% 90% 80% 9% 15% 12% 17% 17% 23% 70% 45% 35% 60% 50% 47% 67% 61% 64% 51% 40% 62% 30% 50% 20% 10% 0% 44% 35% 26% 19% 19% 14% 8% 5% 2% Less than $15,000 $15,000-$24,999 $25,000-$34,999 $35,000-$49,999 $50,000-$74,999 $75,000-$99,999 $100,000-$149,999 $150,000 or more 0 vehicles 1 vehicle 2 vehicles 3 vehicles 4-or-more vehicles SOURCE: U.S. CENSUS BUREAU, AMERICAN COMMUNITY SURVEY 2006-2010 FIVE-YEAR ESTIMATES. SPECIAL TABULATION: CENSUS TRANSPORTATION PLANNING.
Share of Respondents How does vehicle access vary with household income when one groups the households by Metro Corridors? As discussed during Meeting Three, it is difficult to talk about the relationship between income and vehicle availability without also considering a household s access to transit service. Since data that would allow us to look at income, vehicle availability, and transit availability are not available from the Census Bureau, these data come from the 2015 Arlington Resident Transportation study. 1 Note that this survey asked participants to include cars, SUVs, trucks, motorcycles, and other motorized vehicles when reporting the number of vehicles. Vehicle Availability by Annual Household Income, Rosslyn-Ballston Corridor 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 50% 50% Less than $20,000 8% 54% 67% 31% 33% $20,000 - $29,999 $30,000 - $39,999 4% 7% 45% 51% $40,000 - $49,999 62% 31% $50,000 - $59,999 14% 60% 70% 26% $60,000 - $79,999 8% 22% $80,000 - $99,999 21% 16% 60% 68% 15% $100,000 - $119,999 19% 75% 37% 52% 11% 8% $120,000 - $139,999 $140,000 - $159,999 $160,000 - $179,999 38% 41% 53% 48% 3% 6% $180,000 - $199,999 $200,000 or more 0 vehicles 1 2 3 4 5 vehicles SOURCE: ARLINGTON RESIDENT TRANSPORTATION STUDY 2015; NOTE THAT INCOME CATEGORIES AND THE POSSIBLE NUMBER OF VEHICLES AVAILABLE DIFFER FROM CENSUS CATEGORIES BECAUSE OF DIFFERENCES IN HOW QUESTIONS WERE ASKED. 1 This survey collected data on a variety of transportation and demographic topics from 4,008 residents. The survey was designed to get responses from a representative sample of residents. More background on this study can be found in a presentation to the Arlington Transportation Commission on June 30 th, 2016 (audio, video, and further documentation can be found on the Transportation Commission web site here).
Share of Respondents Vehicle Availability by Annual Household Income, Jefferson Davis Corridor (2015) 100% 90% 80% 70% 60% 50% 50% 20% 36% 45% 41% 11% 58% 5% 6% 60% 66% 21% 43% 32% 42% 42% 32% 40% 80% 30% 20% 10% 0% 50% Less than $20,000 $20,000 - $29,999 64% $30,000 - $39,999 55% $40,000 - $49,999 47% $50,000 - $59,999 31% $60,000 - $79,999 34% $80,000 - $99,999 27% $100,000 - $119,999 33% $120,000 - $139,999 61% 4% $140,000 - $159,999 37% 13% $160,000 - $179,999 46% 60% 6% 4% $180,000 - $199,999 $200,000 or more 0 vehicles 1 2 3 4 5 vehicles SOURCE: ARLINGTON RESIDENT TRANSPORTATION STUDY 2015; NOTE THAT INCOME CATEGORIES AND THE POSSIBLE NUMBER OF VEHICLES AVAILABLE DIFFER FROM CENSUS CATEGORIES BECAUSE OF DIFFERENCES IN HOW QUESTIONS WERE ASKED.
Average number of vehicles Here we present the same data but in terms of the average number of vehicles per household. 1.6 Average Number of Vehicles per Household, Jefferson Davis Corridor 1.4 1.2 1 0.8 0.6 0.4 0.2 0 Less than $20,000 $20,000 - $29,999 $30,000 - $39,999 $40,000 - $49,999 $50,000 - $59,999 $60,000 - $79,999 $80,000 - $99,999 $100,000 - $119,999 $120,000 - $139,999 $140,000 - $159,999 $160,000 - $179,999 $180,000 - $199,999 $200,000 or more SOURCE: ARLINGTON RESIDENT TRANSPORTATION STUDY 2015; NOTE THAT INCOME CATEGORIES DIFFER FROM CENSUS CATEGORIES BECAUSE OF DIFFERENCES IN HOW QUESTIONS WERE ASKED.
Average number of vehicles Average Number of Vehicles per Household, Rosslyn-Ballston Corridor 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 Less than $20,000 $20,000 - $29,999 $30,000 - $39,999 $40,000 - $49,999 $50,000 - $59,999 $60,000 - $79,999 $80,000 - $99,999 $100,000 - $119,999 $120,000 - $139,999 $140,000 - $159,999 $160,000 - $179,999 $180,000 - $199,999 $200,000 or more Average Number of Vehicles per Household, Rosslyn-Ballston Corridor SOURCE: ARLINGTON RESIDENT TRANSPORTATION STUDY 2015; NOTE THAT INCOME CATEGORIES DIFFER FROM CENSUS CATEGORIES BECAUSE OF DIFFERENCES IN HOW QUESTIONS WERE ASKED.
How do Arlington residents make their commutes? The 2015 Arlington Resident Transportation Study found the following breakdown of primary commute modes for residents as a whole: Primary Commute Mode for Arlington County Residents Mode Share of Respondents Drive alone (Incl. taxi, Uber/Lyft) 40% Train 27% Bus 12% Bike 5% Walk 4% Carpool/vanpool 3% Telework 3% How does commute mode choice relate to income? Data from a variety of sources reveal that the relationship between income and mode choice is less clear-cut than between income and having a vehicle. The 2013 State of the Commute Survey results for Arlington found that no significant differences were noted for respondents in different age groups or in different income groups. It should be noted, however, that the sample sizes for these sub groups often were quite small (p. 14-15). Preliminary results for the greater region from the 2016 State of the Commute Survey found that [d]ifferences in mode use by income were not clearly defined. Respondents who had incomes in the middle income groups ($60,000 - $119,999) rode a train mode often than did other income groups but use of other modes showed no clear increasing or decreasing patterns by income. (p. 25). The 2015 Arlington Resident Transportation Study provides us with information about residents commute choices based on the annual income that their household receives each year. For the County as a whole, use of driving alone for commuting was approximately the same for all income groups until income reached $200,000 or more. At this level, the drive alone rate was notably higher. The highest income respondents also carpooled and vanpooled at a higher rate and biking/walking was more common among higher income respondents. Bus percentages declined with increasing income. Use of train for commuting was similar across income groups.
When considering Metro Corridors alone, the clearest patterns are in drive-alone commuting and bus commuting; drive-alone commuting is generally more common among individuals with higher incomes (though the trend is not linear) and bus commuting is most common among middle-income commuters. Primary Commute Mode by Annual Household Income in the Two Arlington Metro Corridors SOURCE: ARLINGTON RESIDENT TRANSPORTATION STUDY 2015. NOTE THAT PRIMARY COMMUTE MODE REFERS THE MODE USED MOST DAYS IN A TYPICAL WEEK. FOR AN INDIVIDUAL WHO WORKS FIVE DAYS A WEEK, THIS WOULD BE THE MODE THAT HE OR SHE TAKES AT LEAST THREE DAYS IN A TYPICAL WEEK.
In the table above, the color ranges help depict relationships (or lack of relationships) between income and mode choice. We have not separated these data into the two Metro Corridors because doing so would chop up the sample into groups that are too small to be statistically useful. The survey also had few respondents who were employed and who had annual households with incomes less than $50,000. For this reason, we have not included respondents who reported annual household incomes lower than $50,000 in the color range, but we encourage you to take notice of them. The Sparklines show the relationship between income and mode choice in a different way; the red dot represents the high point.
What Affects Parking Demand? The Park Right DC project Editor s Note: We have arranged for a representative of the District of Columbia Department of Transportation (DDOT) to speak on policy and research undertaken there. The following text describes the Park Right DC project, a recently completed effort that looked at parking at multi-family residential buildings in the District of Columbia. This effort is similar to that undertaken in King County, Washington where Seattle is located. More resources on the King County project can be found at the King County Metro web site here. The text included here comes from a DDOT web site that presents the project. The following paragraph presents an important summary of the study s key takeaways: Generally, as parking supply, average unit size, average number of bedrooms, average rent, and block size increases (or walkability decreases), parking utilization increases. Where there is provision of transit information, greater walk access to transit, greater transit access to jobs, and higher retail and service job density, parking utilization decreases. Background The District Department of Transportation (DDOT) and the Office of Planning (OP) led a research effort starting in February 2014 to understand how parking utilization in multi-family residential buildings is related to neighborhood and building characteristics. The primary goal of the research was to provide a tool to estimate parking utilization on a dynamic website to support and guide parking supply decisions. The intent is that a transparent, data driven process for parking supply decisions may help relieve problems associated with over- or under-supply of parking. The tool relies on local information reflecting residential development and auto ownership patterns in the District. Supported in part by a grant from the Metropolitan Washington Council of Governments, DDOT assembled information about multi-family residential parking use at over 115 buildings covering nearly 18,000 dwelling units in the District during the winter and spring of 2014 and 2015. Parking utilization was recorded on typical weekdays between midnight and 5 a.m. in all residential spaces identified by property managers in each multi-family development studied. Interviews with property managers focused on building characteristics, enabling development of a model that links building characteristics (such as rental price, parking cost, unit size), and neighborhood characteristics (access to transit, density, pedestrian friendliness) to help better match parking demand and supply. What We Learned The entire research process, including coordination with the development community, provided valuable insight to DDOT and OP on factors driving parking supply decisions. On average, in the over 115 developments researched, only 60% of parking stalls are being used. Parking supply was found to be the variable that correlates most with parking utilization accounting with 66% of the variation in observed parking utilization. Other building variables were found to be statistically significant as well, including parking price, average rent, and unit size.
The most significant neighborhood variable was a combination of walkability (measured by block size) and frequency of transit service within walking distance. As walkability and transit frequency increased, parking utilization decreased. The model achieved an R-square of 0.82 indicating that the variables used in the model on average predict about 82 percent of the variance in parking utilization (leaving 18 percent unexplained). This is a very strong model given the complexity of the relationship being researched. The Model The data collected across 92 buildings were used to develop a model of Parking Utilization (observed parked cars per occupied housing unit in the building) at the parcel level. Estimated Parking Utilization from the Park Right DC Model SOURCE: TRANSPORTATION RESEARCH BOARD RODGERS, J., ET. AL. 2016. "ESTIMATING PARKING UTILIZATION IN MULTI-FAMILY RESIDENTIAL BUILDINGS IN WASHINGTON, D.C." PAPER NUMBER 16-4427. TRB 95TH ANNUAL MEETING COMPENDIUM OF PAPERS. WASHINGTON, DC: TRANSPORTATION RESEARCH BOARD.
The independent variables were chosen to optimize both the model s goodness of fit and predictability. The tested variables were grouped into two major categories, variables that describe the building and those that describe the surrounding neighborhood. Variables that describe the building include: Parking Supply Number of stalls provided divided by the total number of units in the building Transit Information Variable equals 1 if transit information is available for building tenants Fraction of Affordable Units Fraction of units set aside for affordable housing Average Unit Size Average square feet for all units in the building Parking Price The average price charged for parking one car in the buildings parking facility Average Bedrooms per Unit Average number of bedrooms per unit for all units in the building Average Rent Average rent for all units in the building Variables that describe the neighborhood include: Block size (walkability measure) Average size of all blocks that intersect a ¼ mile buffer around each parcel Retail & Service Job Density (retail proximity measure) - The number of employees working in these establishments was totaled for establishments within ¼ mile of the parcel. This total is then divided by the land area within this the ¼ mile area. Walkable Transit Trips per Day (access to transit measure) Number of trips available within a ¼ mile for buses and ½ mile for rail using network distances, divided by the area (in acres) within a ¼ mile of the parcel. Jobs by 45 Minute Transit (job accessibility by transit measure) The transit commute time is determined from every block in DC to every Transportation Analysis Zone (TAZ). The numbers of jobs (from the Metropolitan Washington Council of Governments) in the TAZs that are within a 45 transit trip are totaled to create this measure. Development of the regression model considered interactions between the independent variables. For example the Transit Trips per Hour variable was correlated with Parking Utilization, but once walkability (measured by Block Size) and all the other variables were introduced into the regression it was found that the statistical significance was reduced to a level that would not include it in the final model. However, if Transit Trips per Hour and Block Size were interacted then the interaction variable was found to meet the significance criteria. Using this flexible model form has the advantage of finding
significant combinations of independent variables; however, it does make the model somewhat more complicated to understand. The final model is presented below: Variable 1 Variable 2 Incremental R 2 Parking Supply per Unit -- 65.7% Block Size -- 70.1% Transit Information Walkable Transit Trips/Day 72.3% Fraction Affordable Units Jobs by 45 Minute Transit 74.3% Sq. Ft. per Unit Jobs by 45 Minute Transit 76.3% Parking Price Retail/Service Job Density 77.8% Block Size Walkable Transit Trips/Day 78.1% Average Bedroom/Unit Block Size 78.4% Average Bedroom/Unit Sq. Ft. per Units 79.9% Average Rent Retail/Service Job Density 80.2% Retail/Service Job Density Parking Supply per Unit 82.5% Generally, as parking supply, average unit size, average number of bedrooms, average rent, and block size increases (or walkability decreases), parking utilization increases. Where there is provision of transit information, greater walk access to transit, greater transit access to jobs, and higher retail and service job density, parking utilization decreases. DDOT, DCOP, and the consulting team are developing a peer reviewed paper detailing the entire technical process to be posted to this site at a later date.