Sanborn Maps of New Jersey

Apparently, in the time since I last wrote about map resources, the library at Princeton University has made available the Sanborn Fire Insurance Maps for New Jersey.  They're a great resource for seeing town patterns, and the New Jersey patterns are so different than those of the typical Midwestern town.

The older portions of most Midwestern towns are laid out in a regular grid, often aligned with the cardinal directions (North-East-South-West), though topography sometimes alters that arrangement.  In contrast, many of the older New Jersey towns are laid out in a less regular manner - there are fewer right-angle intersections for the major roads.  Those that saw significant growth before the mid 20th century seem to have grid-like additions tying in.  Jersey Shore towns seem to follow the regular grid more than many other Jersey towns.

The link is below for the interested.

http://library.princeton.edu/libraries/firestone/rbsc/aids/sanborn/

Downtown

Downtown. Center City. The Central Business District. Whatever it is called, most American towns of any size and age have an area where most of the businesses were concentrated. Suburbanization, interstates, consolidation, and globalization have often take their tolls on such districts, but most towns continue to have such districts. Usually, such districts are reasonably central to the city. But not always.

Most of the exceptions are found where there is a terrain feature influencing development. Rivers, lakes, and seas can present a formidable obstacle or outright block to extending development in some directions. Hills, mountains, and valleys can have a similar influence. Sometimes even the angle of a slope itself can affect patterns of development.

Take the city of Cincinnati, Ohio for one example. Downtown is roughly centered with respect to east and west, but north and south is another matter entirely. The broad expanse of the Ohio River, beyond which lies the state of Kentucky, meant that any expansion southwards was not part of the city. And hills to the north of downtown were largely restricted to residential development. So downtown Cincinnati exists at the extreme southern edge of the city proper.

Toledo, Ohio is flatter than Cincinnati, but the obstacle presented by the width of the lower Maumee River slowed development to the east. Toledo's downtown is closer to centered than Cincinnati's, but it still isn't precisely balanced.

An extreme example of such a lack of balance in the position of a downtown can be found in the large village of Montpelier, Ohio. As a town of nearly five thousand, Montpelier has a nice little concentrated business district. Despite the relatively-flat terrain, this district is nevertheless located almost at the extreme northwest of the village's development. The vast majority of residential development extends south and east from the downtown. Why? I've got no clue, but it is a bit noticeable when looking at overhead imagery of the area.

Planes, Trains, and Automobiles

I don't know if I've posted about this here before or not, but here goes. The BNSF (Burlington Northern Santa Fe) railroad has track running down the center of Houser Way in Renton, Washington. That's not too unusual, and in fact used to be fairly common, but over the past century or so towns and cities throughout the US have pushed to have railroads reroute their tracks onto private rights-of-way, and as the inner city industries that once required rail service go out of business. So it is a little more rare these days than in decades past. But in this case, it is even more unusual, as the tracks in question carry trains to the Boeing factory. Among the more notable freight those trains carry are the fuselages of Boeing 737 airplanes from a Boeing facility in Kansas to the facility in Renton where final assembly takes place. Linked here is footage of a train carrying several such a load down the street, past parked along the sides of the road and across the route of cars travelling the cross streets. So, we have planes, trains, and automobiles.

Generating Novel Towns

I'm nearing completion on a database and application to gather data regarding American towns from a century ago.  The data collected should give me the details I need to reliably recreate novel yet plausible American-style towns.  However, actually creating the towns requires something else.  It requires the rest of the features I've shown in embryonic form in the spline prototype: roads, railroads, and buildings.

A portion of a Sanborn Fire Insurance Map

So what data do I plan to collect to provide the statistical basis for plausible towns?  There's general data regarding the town that determines some high-level aspects of the town layout: population, the size and shape of the central business district, whether there is a central square, and whether the town is a county seat (which requires a county courthouse).  There's data on the railroad(s) in town that need to be collected, including the spatial relationship of the railroad to the central business district. Data on individual government, religious, commercial, industrial, and utility buildings will be collected, including details of rail service.

Very little data on individual homes will be recorded. An initial glance at Google Earth imagery and Sanborn Fire Insurance maps suggests to me that house style and construction materials have far more to do with date of construction and region than with population or other factors.  Prevalence of Brick vs. stone vs. frame construction varies mostly by region and age of the town, and age of the structure.  Style is dependent upon region and date of construction.    There's also far more data on residential units than the rest of a town, and it is largely less relevant to capturing the feel of a town, in my opinion.

Also collected will be details on businesses and other institutions.  Businesses will include industrial and commercial establishments.  The business data will be tied to the building data.  What's this business data good for?  It lets me know for a given population, how many businesses, and what types, should be added to the town, and where.  While I expect the commercial establishments to vary a little between regions, I expect the industrial establishments to vary to a greater extent.  One would expect more canneries in fruit growing regions and more creameries in dairy regions.  A large town on the American plains should have more grain elevators.  I think a coal yard, lumber yard, and oil dealer could show up anywhere.

So where am I collecting the data from?  Mostly from the Sanborn Fire Insurance maps.  Utah, Indiana, and Pennsylvania all make their pre-1923 maps available online to the general public, I can access the Ohio maps by virtue of being a resident of Ohio, and the Library of Congress has partial online coverage of various states.  USGS Topographical Maps, Google Earth, and historical railroad maps may also play a part.

Screenshot of the prototype

A lot of the other data that is collected will be helpful in procedural generation of novel towns.  Street names, for example, can be pulled from the collected data in a probability-derived manner.  Building footprint size may also be useful when subdividing blocks and generating buildings.  Certainly knowing the number of floors for buildings is useful, as well - especially combined with the population data.  Knowing that, hypothetically, a town with a population of 1000 should have 10 shops along either side of Main Street, with heights of one or two floors, plus three industries along Railroad Street, is useful to make a town believable.

Even after the town data is collected and useful statistics are available, a lot more graphics code will need to be written to actually generate and visualize a new town.  The split-based code I mentioned in yesterday's post is part of the likely building generation solution.  The road and railroad code needs additional work, particularly with respect to intersections and switches, respectively.  And I need to finish up the straight skeleton calculation code, which along with buffer operations (inset/outset) is needed for both building generation and generation of lots for a block.  Speaking of blocks, I need to finish up implementation of minimum cycle basis code based upon David Eberly's paper, which is required to "recognize" the blocks in a generated plan of roads in the town.

All in all, there's still a lot to do, but a little bit more gets done each day.  Until the holidays come. Then work will likely grind to a halt for a few days.  Maybe I can find something else to blog about then.

A Glimpse into the Past: Sanborn Fire Insurance Maps

As I've mentioned previously, I've been making observations about traditional American towns.  I am presently trying to better quantify and systematize the data.  One important source of data are Sanborn Fire Insurance Maps.  The Sanborn company produced maps detailed fire insurance maps of towns and cities across America for over a century.  These maps indicate the buildings in towns, their construction, number of stories, etc. and include roads and conventional railroads (streetcars and electric interurbans are rarely mapped).  The detail they provide helps offer a glimpse into the American past.

A number of additional fire-relevant details are also provided on the maps, as understanding fire risk and fire fighting capabilities was their primary purpose.  Note the water pipes and their diameters shown on the portion of a map shown below.  The black circle with a D.H. above it indicates a double fire hydrant.  Toward the bottom is a symbol with the letters F.E. next to it, indicating a fire escape. More detailed information was provided for industrial facilities and fire stations.

In the portion of map shown, the Great Northern Glove Manufacturing Company occupies the second floor of the structure.  If you look closely, you'll notice that it includes the information that the facility is heated by stove and that power and lights are provided by electricity.

This wasn't always the case.  Some small factories and machine shops in the early 20th century relied upon gasoline engines to power equipment.  In some cases you'll find this noted, including the size of the gas tank(s) in use - but not on this particular bit of map.

From the 1919 Sanborn map of Bluffton, Indiana 

It can take some additional information to help decode the maps.  The maps employed a set of abbreviations for various features and their aspects.  For example, the "Bl. Sm." in the yellow rectangle at bottom right indicates the building is occupied by a blacksmith, while the parenthetical "C. B." on the blue rectangles toward the top indicate concrete block construction.  The number of stories for each building is shown by a somewhat large number in a corner of the building; if a B suffix is present it means there's a basement.  The height in feet is usually shown as well.

The colors of the building as depicted on the map are also important.  They help denote the construction of the building.  The Library of Congress has a nice scan of one such key available. Short version: yellow is frame (wood) construction, pink is brick construction, blue is stone or concrete, and gray is metal - although a shade of gray is also used for adobe.  There are various combinations that indicated combinations of materials, and abbreviations that modify, so a careful look at the key is called for.

So how can you look at these maps?  Digital copies of the maps for a state are usually available free online from a state library and/or university.  What's the catch?  Oftentimes you can only access them from within that state, or with institutional credentials (library or university) from that state.  Partially this is due to copyright.  The ones from 1923 or later are still under copyright, but the ones from 1922 and earlier are in the public domain and may be made freely accessible.  Partially this is due to budget reasons: its costs money to host and stream the maps.

Indiana University and Penn State University have their respective states' pre-1923 Sanborn maps freely available online, but later ones are only available to in-state users.  They're in color.  The Sanborn maps for Ohio are freely available to Ohioans, but not out-of-state - and they're only in black and white.  New Jersey, Illinois, and Iowa seem to be in a similar situation to Ohio, but in color.

The Library of Congress has their digital collection of Sanborn maps accessible to all - but only a tiny fraction of their physical Sanborn map holdings have been digitized.  For example, for New Jersey, only about a dozen cities and towns whose name start with "B" are available digitally - the rest of the alphabet need not apply.  For Iowa, a small portion of the cities and towns whose names begin with "D" or "E" are available digitally.  For some states, no digital Sanborn maps are available from the Library of Congress.

There's also a complete digital collection maintained by ProQuest.  Alas, many institutions don't purchase access.  Invidual access is pricey.

Despite some difficulty gaining full access to the maps, they can be quite useful for learning about the past of towns and cities of America.  Data derived from the maps can be useful in many contexts. It can answer questions about urban growth, patterns of development, mix of stores, and other social, geographic and historic matters.

Alas, as I mentioned above, one are of personal interest where they've proven to be of little help is with respect to streetcars and electric interurban railways.  Mostly, the map makers omitted them, unless they were conversions of steam railroads, or also used steam locomotives regularly.  If they included them otherwise, it appears to have been by lucky coincidence.  Stations and other facilities are generally shown, but the tracks themselves are not.

So, to conclude, the digital Sanborn maps offer a useful glimpse into the past.  You can probably obtain easy access to your state's collection.  Obtaining free access to those for other states may be trickier, but you may be lucky.

Some Observations on Towns

Why make observations about towns?  Because the world is full of them, and if I wish to simulate them, I need to know a bit more about them.  To date, 3D modeling of cities has largely focused on major urban areas, trying to replicate cities such as New York City, Paris, San Francisco, etc.  Only two papers I'm aware of address other types of population centers: there is one paper villages and one on South African informal settlements.  So before I ever attempt to address simulating them, I need to make some observations.  A brief literature search revealed little; most works seem to be on major historical or modern cities, or planning new communities.  I've found virtually nothing on American towns.

What is a town?  For purposes of discussion, we shall label as a town as any population center with a population of several hundred to several thousands, and that has a concentrated commercial district with more than eight commercial structures.  A grocery store, a gas station, a post office, a bar, and a few dozen houses do not constitute a town, but a village.  For this discussion in particular, a traditional town is one whose commercial district predates the mass adoption of the automobile - let us use the date 1930 for the sake of this discussion.  In this discussion, we shall primarily focus on traditional American towns, in particular their shape.

The overall shape, as in the outline of the town, tends to be vary from compact rectangle to amorphous blob.  Sometimes, the main mass of the town extends further out along the roads connecting the town to others, and so appears to have tendril-like growths extending from it.  But that's not the shape I'm really referring to.  Perhaps layout, arrangement, or plan would be better terms.  For example, the shape of the commercial district and the industrial district, and their positions relative to one another, do much to determine the appearance of a town.  The commercial district seems to take three main forms.  I refer to them as linear, intersection, and cluster.

With a linear commerical district, commercial buildings face a single street along one or both sides of the street.  Many times, the street in question bears the name Main, Market, Broad, High, or Commercial.  That street may or may not be the main thoroughfare through town.  In some cases, the street lays perpendicular to the main thoroughfare.  When it is perpendicular, the commercial district may be split by the main thoroughfare, or it may be adjacent.

In an intersection form, the commercial district runs along intersecting streets, often taking a "t" shape, but sometimes a "y" or "k" in towns where the streets take a less grid-like form.  The "y" and "k" shapes seem more common in towns that are East of the Appalachian Mountains, with a "t" shape is more common elsewhere in America, but all of the patterns can be seen throughout the nation.  It is not uncommon for the street names mentioned above to be applied to the streets in question.  As with linear, in the intersection form the streets involved may be the main thoroughfare or not.

The third form I refer to as cluster.  The cluster form consists of multiple blocks of commercial buildings clusterered together.  In a town, this seldom exceeds an area of six blocks by six blocks, and is often far less.  It may be square, rectangular, or somewhat irregular.  Sometimes the commercial district will include a central square that contains a park, courthouse, and/or town hall.  In such cases, the central square is often faced by commercial buildings on all facing half-blocks; in some cases, that is the entire extent of the commercial district.

An industrial district usually runs parallel to railroad tracks, or in older mill towns, along the banks of canals or mill races, where water power could be obtained.  The industrial district is seldom very wide but may run for an extended distance along the railroad tracks or mill race.  If there are multiple railroads or power canals, there may be multiple industrial districts.  This is especially common in cases where one railroad runs east-west and another north-south.

In cases where the town and the railroad where built at roughly the same time, or the town was very small when the railroad arrived, the commercial district is often  adjacent to the industrial district near the railroad.  If the town came first, and grew a bit, the railroad often passed by the edge of town rather than purchasing or condemning the property necessary to get near the commercial district.  In such cases, several blocks of residences may separate the commercial and industrial districts.

And thats all for now.  These are preliminary observations, and I'm still gathering data and organizing it.  Once I've collected and tabulated more data I shall follow up this initial information and quantify things a bit.  Some decent examples and statistics would be nice to share.