GIS and Online Mapping: Stretching the Truth Scale

GIS of Hayes Valley.jpg

When I began my land planning career in 1968, one of the first things I learned about was the use of the Rubber Scale. What is it? Rubber Scale was a term used by civil engineers and land surveyors to describe an inaccurate plan that ignored the physical limitations of the existing terrain. To say that the planner or architect had used a Rubber Scale to create a beautifully rendered plan with pastel colors and soft shadows cast from tree stamps was a negative comment, since these plans were pretty much worthless to the engineer and surveyor that had to make the plans conform to regulations. Because the lines were hand drawn back then (and in many cases still are today), accuracy was, and remains, an issue.

I was guilty of “stretching the scale” to maximize density, thus the term. The rubber scale was beloved by designers who wanted to look good to their developer clients. The developer expected a plan that maximized yield, and a plan that was of a higher density than they expected would surely be pleasing. Of course, these plans had no basis in reality. Ultimately, the land surveyors and civil engineers would lose the units we falsely claimed, and they would also get blamed for the density loss!

Fast forward two decades to 1988, when GIS (Geographic Information Systems) began gaining market share. Government agencies (typically cities and counties) embarked on spending sprees with the promise of a new era in planning technology brought on by the advent and proliferation of the GIS technology, which blended graphic mapping information with a data base. You wanted to know property information, demographics, soil types? Just query the map. The sales teams of GIS mapping systems convinced those in charge of purchasing that “parcels” shown on the map could be traced quickly from a variety of sources, then later “rubber sheeted” into accurate surveyed section corners. As if by magic, inaccurate parcels would be made precise.

So— early, existing hand drawn maps were traced into a computer, and the imprecise data was made even more imprecise. Even when the data came from aerial maps, created by flights several thousand feet above the surface, the accuracy was at best within three to five feet of actual location. What happens to curved boundary lines if the map is stretched to meet tens of thousands (or more) of lot corners, with all four section corners set accurately? The answer is, of course, a map that would be impractical to correct at a later date. Very few GIS maps exist today that were done using accurate land survey from the beginning.

Fast forward another two decades and more to today, when Google Earth and its rivals, MapQuest and BingMaps, have unfortunately become the basis for site information. Call the data of these suppliers “on-line graphics”. There are a variety of software systems that boast that site layout can easily be done using on-line-graphics, or simply using the available on-line GIS mapping data.

Here lies today's problem: None of this information is likely to be accurate enough to be useful to an engineer or surveyor who ultimately must put their license on the drawing, guaranteeing its accuracy.

GIS salesman make their sales commissions by convincing the world's governments that data can be “adjusted” later to a more accurate data structure. This is true, but not economically feasible or practical. Since curved property lines are represented in a GIS system by a series of miniscule lines, it could be possible for the data to be corrected. This is like saying that it could be possible to temporarily build an approximate building and later on move walls to the locations shown on the plan. It would be possible, yes, but hardy cost (and time) effective.

A lack of correct information is a huge problem when using GIS or on-line-graphics information. For example, it’s not possible to see contour lines that are accurate enough to determine flood plain, wetlands, or other information critical to the initial site design. Even when this information is shown on the city or county on-line map, what is the source of that data? Most of these maps are sourced to the lowest bidder. Was that wetland shown on the site just something that looked wet on an aerial map traced by a low-bidding draftsman, or was the wetland defined by an environmental expert who accurately surveyed it and somehow placed it precisely on the GIS map? This is the modern day version of using a rubber scale to measure the very data used to make decisions. Entire cities are stretched beyond practical use by surveyors and engineers.

Many think that the on-line-graphics are taken from a satellite in space with some military camera that spy agencies use. Wrong! The images are derived from aerial mapping firms. These photographs might be several years old, which is why you might look at a newly developed suburban area, while the map still shows a farm field. You could be looking at a site to build residential units, not realizing a sewage treatment plant was recently built next door.

Building Information Modeling (BIM) is a way for architects, engineers, contractors and owners to communicate and collaborate while designing a structure they are all involved in creating. The various parties can be instantly communicating worldwide on a single project located anywhere. Recently, I was participating in an on-line conference from one of the suppliers of BIM technology, who wanted to sell me on the idea of using BIM as an additional tool to use with our precision site design software. I require a developer to furnish us with an accurately surveyed boundary, topography, and any wetlands delineated precisely on the site before I begin my work. The BIM supplier took an office building from Florida (created accurately), and placed it on a site in South Dakota shown by Google Maps as if somehow that’s all there is to planning. Nowhere in the conversation did the salesman mention local regulations, site restrictions, where any easements might be (most often easements are not easily seen in photographs), etc. As soon as I was shown how a building from Florida could be placed on a site in a northern state using an on-line-graphics data base as the planning solution, the demonstration was over. Could BIM be used for site design? Absolutely, but only by using precision data from qualified engineers and surveyors, not on-line-graphics.

The tax payers have financed billions of dollars worth of GIS systems with map data of questionable accuracy, with the understanding that the rough mapping data could be rectified accurately later. While true, the cost of converting existing maps would be prohibitive. The general public might think that these public data structures replace the need for land surveys and accurate civil engineering. But these vague images actually make it even more important to consult with a licensed professional to provide precision data before any design or development decisions are made.

Over four decades ago I was guilty of using the rubber scale. New technology has promoted us into a new problem, and moved us from a rubber scale to a rubber sheeted world.

Photo by edibleoffice: GIS of Hayes Valley, San Francisco.

Rick Harrison is President of Rick Harrison Site Design Studio and Neighborhood Innovations, LLC. He is author of Prefurbia: Reinventing The Suburbs From Disdainable To Sustainable and creator of Performance Planning System. His websites are rhsdplanning.com and performanceplanningsystem.com.



















Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I trust the technology today

I trust the technology today will solve the project limitations, I don't think drawing lines by hand will be an issue for a long time. Take a look at the pipeline alignment sheets software and you'll understand better what I am trying to say.

Missed the mark again Mr. Harrison

First of all, let me say, as someone who has a masters in urban and regional planning and have sat through too many GIS classes, that I am not a fan of GIS and have many, many criticisms of the technology and how it is used. Unfortunately, Mr Harrison again misses the mark when trying to criticize another element of planning that he does not understand.

NO ONE, and I mean NO ONE who has any practical knowledge of GIS would recommend using the information for surveying purposes. In fact, most of the maps have disclaimers explaining the scale and purpose. The GIS in my office would tell/ask you "we are not surveyors, why would you use the information that way?"

Mr. Harrison GIS certainly has its problems ... sometimes people let the tail wag the dog and projects become GIS driven. However, the real purpose and use of GIS in the real world (not the libertarian lala land you exist in) is to simply assist the intuition of the planner.

It's not the GIS, it's the data

The article’s main point that authoritative geospatial data should be created and certified by Professional Land Surveyors is one that I have advocated throughout my 35+ years as a PLS. However, the implication that GIS software is not capable of storing precise and accurate surveying-grade data is not correct. The statement that “… curved property lines are represented in a GIS system by a series of miniscule lines…” is not true for ESRI or Intergraph GeoMedia software that both store curves (simple, Bezier, etc.) as mathematically-defined objects in relational database systems (i.e., Microsoft SQL Server or Oracle). The only ESRI data formats that store curves as densified line segments are shapefiles and coverages, which are dinosaurs in modern enterprise GIS deployments. Another statement, “A lack of correct information is a huge problem when using GIS or on-line-graphics information”, seems to imply that GIS data is not ever capable of being used for engineering-grade applications. If the stored data itself is inaccurate, then garbage in, garbage out. But any modern-day GIS can store numeric quantities to more than enough digits of precision to support accurate data analysis. The key to obtaining useful accuracy is employing professionals who have the expertise to collect and manage geospatial data, and also to describe its provenance through adequately-documented metadata. Whether that data resides in spreadsheets, CAD drawings or GIS’s is largely irrelevant (except when it comes to the ultimate usefulness of that data, which is arbitrated by the… users).

An excellent example of an accurate, survey-grade GIS known to this commenter is the City of Tempe (AZ) TGIS system, in use since 1988. Its Oracle geodatabase was COGO-constructed from original plats, tied to a Second Order Class II network of PLSS quarter-section points. Because of its accuracy, it has been used daily as an engineering-grade design data source.

Rudy Stricklan, RLS, GISP
Mapping Automation, LLC

Another example of the

Another example of the misuse of GIS mapping is by the BC Assessment Authority in British Columbia, the authority that assesses real property which property taxes are based on. In my example, the BCAA digitized a lake waterfront property to determine the area. The property title was based on a very old plan (early 1900's)and the only lineal dimension was along the road fronting the property. The difference between the natural boundary and the high water mark was significant due the approximately 100 years that have passed since the first survey (the bed of the lake is owned by a private entity thus the rules of acretion/erosion etc. don't apply). The natural boundary is very sinuous and if I was asked to 'upgrade' the title by creating a reference plan to be registered in the land title office I would re-establish the road boundary and then scale the natural boundary using the plan that the title was based on not the local government GIS mapping. The difference in the area using the scaled-from-the-title-plan method and the GIS method was significant; 0.8 acres for the correct method and 1.2 acres for the GIS method. When this was pointed out to the BCAA appraisers the only concession given to my client was a reduction in their assessment based on o.8 acres v 1.2 acres. However, I did not get an acknowledement by the BCAA that their methods were wrong and that they would change their way of determining areas thus, the problem persists.
R. D. Wright, BCLS, CLS

The End User Has Changed

I do not believe that the data produced for the city/county GIS was ever really intended to be used by engineers and surveyors in the first place. I think that most of the intended users at the city/county level WERE informed and aware of the limitations of the data that they were creating or had been provided. This was not a problem when GIS was something that only GIS professionals had access to. Most had no need to go to the even larger expense of ensuring that all data created was survey accurate because they didn't need the data to be survey accurate for their intended use. But the popularity of online, freely and readily available web mapping programs have put GIS into the hands of end users who were not the originally intended end users when the data was first created. Many are end users who expect too much out of the data that they have access to, and are not adequately informed of the limitations that this data possesses.

GIS data

Rick: Like a lot of things - this is an "education" issue. The General Public needs to be educated on the limitations of GIS as it relates to surveying/property corners. This is no different than the little "go round" that I had with an article in the StarTribune back in 2007. A couple had written to the Strib complaining that their County would not give them coordinates for their property corners so the couple could go out and find/locate their property corners! The writer for the Strib referred them to Google Maps and told them to go there and pick off their coordinates from that mapping data base and then take their handheld GPS and go to those coordinates and that would be the location of their corners! The writer, the individual from Google Maps, the couple, AND the General Public do not understand surveying and proper property corner placement. A section is NOT 5280 feet by 5280 feet and it does not have 90 degree corners and it does not contain exactly 640 acres. This is part of the "education" that needs to be done. The other part is to explain that the aerial photo "does not lie", BUT the placement of any "property lines" on that aerial photo is not "survey accurate", and therefore cannot be relied upon to pull coordinate values from the photo.

Tom Veenker
Aitkin County Surveyor
Aitkin, Minnesota

The Right Tool

Great article. We can better serve our clients and the public by using all of the tools available with the critical understanding of when and how to use them. Hiring licensed professionals to do the research, planning, surveying, designing and engineering for any project is paramount. It is our duty as professionals in this field to use the right tool at the right time.

David Horn, PE, LS
Yamabe & Horn Engineering, Inc.
Civil Engineering - Land Surveyoring - GIS
Fresno, CA

GIS and Online Mapping

Rick does a great job of outing the side of GIS that most will fail to discuss. Vast amounts of money are spent on GIS systems that in the end are destined to disapoint or harm the end users. It is time to realize that such a great tool should not be built on a weak foundation.

Ray Niles, PSM