Designing a Mapping Project

Author: Brian Mayfield, Director of Proposal Development / GIS Mapping Scientist
brianm@surdex.com

Abstract: Recent market analyses indicate that the number of counties and municipalities that have implemented an enterprise GIS for their community has grown exponentially in the last five years. Increasingly, these communities are turning to the photogrammetric industry to provide accurate and precise mapping data for use in their enterprise system. This paper examines some of the considerations a community must give when developing a mapping project.

Introduction

Surdex Corporation has served County, Municipal, State, Federal and Private Agencies since 1954. During that period, we have been recognized as a premier geospatial data provider; supplying accurate and precise information to our customers - on time and within budget. We attribute our longevity and success to an unparalleled understanding of our clients' needs and goals. Unlike many of our competitors, our objective is to help you meet your goals - accurately, on time, and within budget.

The following sections should provide a starting point for the development of a county or municipal mapping project. Essential elements of a mapping project such as selecting a final product accuracy, photo scale, and contour intervals are discussed hereafter.

Identify Your Needs

The uses of a mapping product often dictate the type of data that will be needed and the accuracy of the data that must be obtained. The first step in developing a mapping project is to identify all current projects and to forecast any potential projects that might use the data produced for a mapping project by a photogrammetric vendor. This is called a needs assessment.

A Needs Assessment may range from a rather informal process of brainstorming with a few department heads to a very formal analysis conducted by a consulting firm; regardless of how you approach the topic, conducting a Needs Assessment is the single most important phase in planning a mapping project and releasing an RFP. Leaping directly to mapping scales, product accuracies, photo scales, pixel resolutions and so on puts the cart before the horse: the result may be highly accurate and precise data, but data (you discover down the road) that doesn't do exactly what you'd like it to do, or doesn't serve the needs of everyone in the Community.

Illustrated on the following page are some general questions that you can ask yourself and others involved in the development of a mapping project for your community. These questions are very basic and should be sufficient to provide a generalized direction for developing a mapping project that meets the needs of your community.

Choosing the Output Map Scale

Choosing the output map scale is perhaps the most important component of designing a mapping project. The output map scale is the first consideration and perhaps the most important specification when designing a mapping project. The output scale determines the size of the output map for a defined geographic area and most importantly, it determines the amount of detail that can be represented on or extrapolated from the map. Like accuracy, the output map scale greatly affects project costs and scheduling. The larger scale map that is required (i.e. 1"=100' is larger than 1"=200') by a community, the longer it will take to produce and the more costly it will be.

The products that are needed by your community as determined in your needs assessment will dictate the map scale to be produced. Many features like street centerlines, edge of pavement and buildings can be captured from many different map scales ranging from 1"=50' to 1"=400'. Typical map scales for municipal mapping applications are 1"=100' for urban or developed areas and 1"=200' or 1"=400' for rural and less developed areas. However, should it be determined that features like fire hydrants or manholes are needed by the users, your community should consider an alternative such as 1"=50' or larger. It is important to select a mapping scale that ensures you can identify each feature you want to collect. Equally important is not to procure a scale that costs more without any practical benefit. If your community is unsure what types of features can be captured from each output mapping scale, please consult a photogrammetrist.

Product Accuracy

Accuracy standards vary in complexity and usability. The most commonly used data accuracy standards for county and municipal are the American Society of Photogrammetry and Remote Sensing (ASPRS) Class I and II. Additionally, more and more counties and municipalities are requesting their mapping projects to be compliant with the National Map Accuracy Standards (NMAS) for large-scale mapping.

NMAS generally equates to ASPRS Class 1.5. Members and clients of the photogrammetric community often misunderstand the United States National Map Accuracy Standards. Often clients ask for products to meet NMAS without clearly stating their interpretation of NMAS. Surdex recommends that each client carefully examine each potential vendors interpretation of the NMAS. By not doing so, he or she could potentially receive a mapping product that does not meet the product accuracy expected. Our understanding of NMAS, which is based on the United States Army Corps of Engineers' interpretation, is as follows:

NMAS, as defined by the United States government, states that the random error for 90% of the population is within 1/30 of an inch at publication scale, which is 6.67' for a 1"=200' map. To relate NMAS in terms of limiting RMSExy, which is the horizontal product accuracy of a photogrammetric mapping product, the following assumptions must be made:

Since the random errors or discrepancies are distributed about a zero mean, the 90% error is also called the Circular Error (CE). To compute the "circular map accuracy standards" (CMAS), which corresponds to 90% of the population: CE = 2.146' * RMSExy or more simply RMSExy = CE / 2.146'. So, RMSExy = 6.67' / 2.146', which is equal to 3.108' or approximately half way between ASPRS Class I (2.0') and ASPRS Class II (4.0') for a 1"=200' map, thus we term NMAS to be approximately ASPRS Class 1.5.

The following tabular information is excerpted from the USACE Engineering and Design Manual for Photogrammetric Production. It illustrates the Limiting horizontal Root Mean Square Error (RMSE) for each class of accuracy as it relates to mapping scale.

Target Map Scale

Limiting RMSEx,y (Values in Feet)

1in = x ft

Ratio, ft/ft

Class I (ASPRS)

Class 1.5 (NMAS)

Class II (ASPRS)

50

1:600

0.5

0.75

1.0

100

1:1,200

1.0

1.5

2.0

200

1:2,400

2.0

3.0

4.0

400

1:4,800

4.0

6.0

8.0

The map scales listed above are indicative of the most popular map scales requested by counties and municipalities for their mapping projects. The RMSE is the square root of the quotient of the sum of the squares of the errors divided by the number of measurements. In other words, the 1"=100' scale mapping compliant with National Map Accuracy Standards would have a horizontal accuracy greater than or equal to +/- 1.5 feet.

The product accuracy also significantly increases project costs and schedules. Many counties and municipalities often fall short of meeting their goals by assigning the wrong product accuracy specifications to their project. The product accuracy should be a derivative of your communities mapping needs and budget constraints. They falter by assigning a very strict product accuracy (e.g. ASPRS Class I), thus limiting the amount of mapping data they can receive within their budget.

Choosing the Photo Scale

Choosing the correct photo scale, which is the flying height (Above Ground Level) divided by the camera's focal length is the key to a successful mapping project. Surdex has performed thousands of aerial photography missions during our 47 plus years of experience. This has provided us with an intimate knowledge of the limitations of photographic and photogrammetric equipment.

Surdex works regularly with the US Army Corps of Engineers to test and recommend mapping specifications. The first table listed illustrates the minimum negative scale (or photo scale) needed for topographic development. The table, which is excerpted from the USACE Engineering and Design Manual for Photogrammetric Production, shows the Contour Interval in Feet and the values that would need to be used to achieve a product accuracy of ASPRS Class I and II as well as NMAS.

Contour Interval (ft) ASPRS Class I NMAS ASPRS Class II
½ 1"=167' 1"=175' 1"=183'
1 1"=333' 1"=350' 1"=367'
2 1"=667' 1"=700' 1"=733'
4 1"=1,333' 1"=1,400' 1"=1,467'

As you can see, the contour interval will greatly affect the photo scale. Also, the lower photo scale will result in an increased cost for aerial photography acquisition and will greatly increase the minimum amount of models that will need to bet set up for photogrammetric compilation of the Digital Terrain Model and the requested planimetric features. When combined, these factors will increase the project schedule and cost.

When constrained by budget, many counties or municipalities decide to acquire digital orthophotography and/or planimetric features only. In this case, we refer to another table from the USACE Engineering and Design Manual for Photogrammetric Production. This table illustrates the minimum negative scales (photo scale) needed for Digital Elevation Model Extraction and Planimetric Mapping.

Target Scale1" = x ft. ASPRS Class I NMAS ASPRS Class II
50 1"=350' 1"=375' 1"=400'
100 1"=700' 1"=750' 1"=800'
200 1"=1,400' 1"=1,500' 1"=1,600'
400 1"=2,800' 1"=3,000' 1"=3,200'

As you can see, the overall project cost would be reduced significantly by eliminating the contour component from the mapping. The most important item when selecting mapping features to be produced for a project is to take careful examination of usefulness and necessity of certain types of data such as 1' contours. More information about the selection of the appropriate contour interval for various applications will be further discussed in this document.

Control and Aerial Triangulation

Surdex regularly works with our clients to utilize existing ground control for their mapping projects. When acceptable, Surdex utilizes Airborne GPS (ABGPS) to further densify the control network. By allowing a vendor to utilize ABGPS technologies to supplant ground control, a county or municipality can realize a significant cost and timesavings.

As illustrated in the table below, which is excerpted from the USACE Engineering and Design Manual for Photogrammetric Production, the survey accuracy requirements are a direct reflection of the mapping accuracy standards chosen and the contour interval of the final mapping product.

Topographic Points for Class Spot or DTM Elevation Points for Class
Target CI (ft) ASPRS I NMAS ASPRS II ASPRS I NMAS ASPRS II
0.5 0.17 0.25 0.33 0.08 0.12 0.16
1 0.33 0.495 0.66 0.17 0.25 0.33
2 0.67 1.0 1.33 0.33 0.5 0.67
4 1.33 2.0 2.67 0.67 1.0 1.33
5 1.67 2.5 3.33 0.83 1.25 1.67

This table further illustrates the importance of choosing the correct components for your communities mapping project.

Fully Analytical Aerial Triangulation

For over three decades, Surdex has been perfecting the science of Fully Analytical Aerial Triangulation. All of our FAAT is performed in-house using only First-Order Fully Analytical SoftPlotter Instruments and experienced technicians. Surdex has long been a recognized leader in the science of FAAT, providing this service to other photogrammetric firms as well as to government entities and private agencies.

The purpose of aerial triangulation in the photogrammetric production process is to establish precise and accurate relationships between the individual photographic film coordinate systems and a defined datum and projection. This relationship is used to link the ground surveyed control points via photographic measurements. The result of the triangulation is a densified set of ground control points that are used to control the remainder of the mapping process.

The maximum allowable error for the triangulation process is demonstrated in the following table, which was excerpted from the USACE Engineering and Design Manual for Photogrammetric Production.

Accuracy Method RMSEx,y Maximumx,y RMSEz Maximumz
ASPRS I Fully Analytical H/10,000 3 RMSE H/9,000 3 RMSE
NMAS Fully Analytical H/9,000 3 RMSE H/7,500 3 RMSE
ASPRS II Fully Analytical H/8,000 3 RMSE H/6,000 3 RMSE

H represents flying height (above ground level) in the table above. These computations provide us with the maximum allowable error for the aerial triangulation process, making it easy to identify errors or deviations from the desired product accuracy.

DTM or DEM Collection

A DTM (Digital Terrain Model), which is needed for the generation of contours, is a highly accurate representation of ground surface using mass points and breaklines. A DEM (Digital Elevation Model), which required for the generation of digital orthophotography, is a less accurate representation of the ground surface using a regularly spaced grid of mass points and breaklines. Surdex regularly compiles both types of surfaces for county and municipal projects. However, Surdex will not compile a DEM in addition to a DTM if the project deliverables include contours. We will simply use the DTM in the development of the digital orthophotography.

The compilation process for a DTM or DEM involves the collection of a dense pattern of mass points and breaklines in the stereo model. The required mapping product (scale and accuracy) determines the placement density of mass points regardless of whether the surface is a DEM or DTM. Breaklines are collected along points of inflection in the topographic surface of the earth, i.e., places where there are sharp changes in the direction of slope on the earth's surface. Some examples where breaklines are placed are at the edge of road surfaces, bottom of creek beds, along hydrographic features and along the top of ridgelines.

Contours

The tables shown in the previous sections illustrate how the selection of the appropriate contour interval for a countywide project will greatly influence the overall project cost and schedule. The following table is excerpted from the USACE Engineering and Design Manual for Photogrammetric Production. It provides the recommended use for contours. Surdex recommends that your community closely evaluate the purposes of each contour interval as listed below and choose the contour interval that is appropriate for your mapping needs as determined during the needs assessment.

Contour Int. (ft) General Purpose
1 Final design, excavation and grading plans, earthwork computations for bid estimates, and contract measurement and payment.
2 Route location, preliminary alignment and design.
4-5 Preliminary project planning, hydraulic sections, rough earthwork estimates.
10-20 High-gradient terrain, low unit cost earthwork excavation estimates.

As previously mentioned, the contour interval will significantly affect the cost and schedule of the project. It along with product accuracy will greatly influence photo scale, which will affect the cost and scheduling of aerial photography acquisition, stereo compilation, and quality control.

The following table reiterates how the contour interval affects the overall project cost and schedule:

Contour Interval (ft) ASPRS Class I NMAS ASPRS Class II
½ 1"=167' 1"=175' 1"=183'
1 1"=333' 1"=350' 1"=367'
2 1"=667' 1"=700' 1"=733'
4 1"=1,333' 1"=1,400' 1"=1,467'

As previously mentioned, the lower photo scale will result in an increased cost for aerial photography acquisition and will greatly increase the minimum amount of models that will need to bet set up for photogrammetric compilation of the Digital Terrain Model and the requested planimetric features.

Summary

The photogrammetrist's ultimate goal is to achieve the highest degree of precision and accuracy possible, providing the end user with the most reliable data possible. Our hope is that this paper will lend itself to help in the development process for a county or municipal mapping project.

Work Cited

US Army Corps of Engineers. 1993. Engineering and Design Photogrammetric Mapping.
Engineer Manual
No. 1110-1000.


Brian Mayfield
Director of Proposal Development / GIS Mapping Scientist
520 Spirit of St. Louis Boulevard
Chesterfield, Missouri 63005 (greater St. Louis metro area)
Voice: 636) 532-3427
Fax: (636) 537-9638
E-mail: brianm@surdex.com

Surdex Corporation has been recognized as a premier geo-spatial data provider since 1954, supplying accurate and precise information to a variety of clients from coast to coast in a timely and economical manner. Surdex provides geo-spatial services for municipal, county, state, federal, and private agencies interested in the acquisition of accurate geo-spatial data, GIS applications, and enterprise solutions for their respective markets.