Request For Proposals Specifications Key To Success

This is the second article in a series about writing good Request for Proposals (RFPs) to ensure high quality geospatial services. In the previous issue, we discussed some of the advantages and disadvantaged of RFPs and how to avoid some of the typical mistakes made by requesting incomplete information. In this edition, we will address geospatial specifications.

No other item effects geospatial RFPs more than positional accuracy specifications. Buyers often leave out or inadequately define these important measures. When not clearly specified, it may be difficult or impossible to compare approaches and costs between contractors. Such lack of specificity results in sub-standard services.

What Are Positional Accuracy Standards?

Positional accuracy is typically the single most expensive component of your geospatial project. During the initial project planning stages, it is important to understand how users will interact with and use the data. This, in turn, will dictate what horizontal and vertical positional accuracies are needed and what can be afforded. Since the cost of positional accuracy increases with accuracy, understanding the needs versus wants of the user community is important to avoid excessive cost (and accuracy) and maximize value.

Guidance from a professional geospatial provider is important at this stage both to grasp the relevant user issues that have a bearing on needed positional accuracy and to determine how much accuracy is affordable. Misjudgment in how much (or how little) accuracy is required and poor project design are the most common reasons geospatial projects fail. A contractor selection process maximizing relevant professional expertise will also maximize the value of the procurement for you and your user community.

It is important the RFP state a certified photogrammetrist on staff is a requirement for all bidders.
These professionals receive national certification from the American Society for Photogrammetry and Remote Sensing (ASPRS) every three years per educational/experience requirements. This professional credential maximizes the odds the professional has a thorough understanding of positional accuracy specifications and mapping project design.

Understand the Positional Accuracy Standards

Understanding the different positional accuracy specifications will help you discuss your project. The three most common specifications include:

• National Map Accuracy Specifications (NMAS)
• American Society of Photogrammetric and Remote Sensing Specifications 1990 (ASPRS)
• National Standard for Spatial Data Accuracy (NSSDA)

Written when all mapping was performed on paper, the National Map Accuracy Specifications (NMAS) fit the era of hardcopy paper and fixed-scale mapping. Later, the American Society of Photogrammetric and Remote Sensing (ASPRS) 1990 specifications rewrote the NMAS and adapted them to map services delivered as digital files, where the scaling of data occurs dynamically because of its digital nature. The most recent mapping specifications, the National Standard for Spatial Data Accuracy (NSSDA), provide a common scale-independent statistical measure of accuracy allowing users to compare datasets across applications.

The table below summarizes how the NMAS & ASPRS positional accuracy standards compare with each other. Keep in mind, because some of these are older standards, photogrammetrists using modern photogrammetric equipment (such as analytical and softcopy stereoplotters, autocorrelation, gyro-stabilized aerial camera mounts, digital cameras, and forward-motion compensation) may be able relax these conservative guidelines and still achieve the desired spatial accuracy.

Comparison of Horizontal Accuracy Standards: NMAS & ASPRS1
Map Scale	Representative Scale	Horizontal RMSE (Feet)
		ASPRS Class 1	NMAS Class 1.5	ASPRS Class 2
1″=50′	1:600	0.5	0.75	1.0
1″=100′	1:1,200	1.0	1.5	2.0
1″=200′	1:2,400	2.0	3.0	4.0
1″=400′	1:4,800	4.0	6.0	8.0
The maps scales shown here are commonly used for large scale mapping projects. So these figures indicate, for example, that 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 Federal Geographic Data Committee (FGDC) developed the NSSDA positional accuracy specifications to provide a common reporting mechanism so users can directly compare datasets. NSSDA uses a statistical estimate to measure positional accuracy called RMSE (root mean square error). Accuracy is reported in ground units at the 95% confidence level. This means 95% of the positions in the dataset will have an error with respect to true ground position that is equal to or smaller than the reported accuracy value. The reported accuracy value reflects all uncertainties of ground coordinate values in the product. For example, if the buyer decides they want orthophotography created with a horizontal accuracy of ±1.0′, then the photogrammetrist would deliver orthophotography where 95% of the dataset will have a horizontal positional error =1.0′. The beauty of the NSSDA standard is it allows any positional accuracy to be specified and defines statistical testing measures to ensure this positional accuracy is achieved to a defined level of confidence.

All three accuracy specifications describe how to measure and statistically estimate the positional accuracy of data. Generally, this procedure requires measuring a minimum of 20 well-defined checkpoints distributed to reflect the geographic area mapped and estimate the error in the dataset. Valid checkpoints can be located in the field or in any other mapping datasets with higher accuracies. Only checkpoints positioned on well-defined features like road intersections, sidewalks, targets, etc. are useful for statistical verification of positional accuracy.

Field verification is not often performed because it is expensive and may not be required if the project is executed by qualified professionals. In these cases, the mapping is delivered with a statement indicating it was “compiled to meet XX units horizontal accuracy at 95% confidence level and YY units vertical accuracy at the 95% confidence level.” Only certified photogrammetrist are qualified to make such claims.

Specifying Positional Accuracy

The specific accuracy standard specified in your RFP is less important than defining the spatial accuracy for each given deliverable. Consider required spatial accuracy carefully based on the anticipated needs of user community. Spatial accuracy is expensive, so identify accuracy needs not wants. There normally will be a tension between accuracy and budget. By balancing spatial accuracy needs with budget realities, you can provide the best value. Your professional geospatial service provider can also help you identify the best value using current geospatial technology.

As mentioned, it is important RFPs specify the required spatial accuracy for deliverables. If not, the photogrammetric consultants will not have a common baseline for making the important decisions about scale, precision, detail, and cost. The result for the buyer will be proposals with such widely varying positional accuracy and costs that it will be difficult or impossible to realistically compare the value of each proposed project approach.

Then, Leave “How To” To Provider

The RFP ideally should only specify the desired spatial accuracies. For instance, the photo scale, flight height, and equipment should generally not be dictated by you. Leaving this control with the geospatial professionals allow them to maximize your value. This has become increasingly important as digital sensors and airborne GPS technology have come to market. These technologies enable achieving specific positional accuracies using vastly different project parameters than could be done historically with analog (film) equipment.

Set your required spatial accuracies, insist on receiving references, and request previous project experience, but leave the “how to achieve your spatial accuracy” to the vendor.

Government Cooperation, More Value

Public agencies considering procurement of digital mapping services should consider cost sharing with other government sources. Federal agencies often need quality data and will share acquisition or conversion cost if the data can be standardized to benefit all users. In fact, because the Federal government insists on standardized data, the mapping data is often more useful to more user groups than envisioned.

An example of such a program is FEMA’s digital flood insurance rate maps (DFIRM). The agency has established a cost-sharing program through its Cooperating Technical Partners (CTP) for local governments (www.fema.gov/plan/prevent/fhm/ctp_main.shtm). The National States Geographic Information Council (NSGIC) is also promoting the Imagery for the Nation program to facilitate data/cost sharing for geospatial services. The USDA APFO has been sponsoring Federal/State cooperative projects for several years through its National Aerial Imagery Program (NAIP).

Positional Specifications Lead To Best Value

By understanding positional accuracy specifications and including them in your next request for proposal, you can be sure your project data is accurate and represents the best value possible. Then, understand the user communities’ needs and balance them with budget realities. If you are in doubt, ask for help from a trusted certified photogrammetrist.

Look For Part 3 In The Next Issue

In the Part 3 of this RFP series, we will discuss how to indicate proper use of ground control. Learn how these processes work and what aspects are important to mention in Part 3 of our RFP series.

Questions Or Comments?

If you have questions about this article, feel free to contact Aerial Services, Inc.’s experts for advice with no obligation. We are here to help you.

References

1. USACE Engineering and Design Manual for Photogrammetric Production