How well do you know your GIS data?

April 2013 » Columns » CE + GIS INTERSECT
Janet Jackson, GISP

GIS perspective
Jackson:
Usually the answer is "not well enough," but if your answer is "really well," then you are fortunate. But can you easily answer questions about using your GIS data for modeling purposes? Knowing your GIS data's strengths and weakness is called data confidence. And being confident in your data doesn't mean you are ready to use it for modeling; it means you are aware of your data's accuracy and completeness. These are great first steps when using GIS data for modeling or other applications.

Not all data is created equal. That is especially true in the case of getting and keeping your data in model-quality condition. It doesn't just happen; it takes months, if not years, of constant and diligent effort by everyone who uses the GIS data. The GIS technician who records the data, makes the updates, and confirms the changes with the field crews is an important member of the team, but is only one part of data entry and quality control. Third-party data providers (developers and consultants) are often key providers of new or proposed data, so it's important that lines of communication are open between them and the GIS group. They are often the first to bring field versus GIS-data-quality errors to the attention of the team so changes can be made.

Other key members of the GIS team are the county or state agencies that gather, sort, and filter data. All these sources and people are constantly changing and creating more challenges and obstacles to getting and keeping model-quality data.

Despite obstacles, all of the effort is worth the payoff when the GIS data is put into a model (sometimes with a few assumptions noted in the metadata) and you see a wide variety of what-if scenarios come to life. Remember: Model-quality data doesn't just happen; it takes time, effort, and lots of team members who know that the more you know about GIS and the GIS data, the better off everyone will be.

Janet Jackson, GISP, is president of INTERSECT (www.intersectgis.com), a GIS consulting firm. She travels the country talking about the importance of intersecting GIS with other professions to create effective solutions for clients. She can be contacted at jjackson@intersectgis.com.

Civil engineer perspective
Mitchell:
The application of GIS data to hydrologic and hydraulic modeling has expedited the overall production of system models via readily available mapping, attribute tables and spatial references. Several modeling software packages have routines that will allow the engineer to populate the attributes required for an operational model, and this perceived "easy" task can often catch an unsuspecting owner and engineer off-guard.

The system owner may have had a long-term GIS staff member in place that created the initial GIS from as-built drawings and staff institutional knowledge; therefore the owner anticipates the data to be adequate for use in modeling and design. As Jackson said, owners may even declare the health of their data is "really well," but has the data been scoured to the detail required for modeling or design, or is it just "really well" for plotting maps or overall planning discussions?

Modeling or design engineers charged with producing a deliverable while using this "really well" data must use significant caution or fall to the pitfalls of producing an inaccurate deliverable. Everyone wants to believe the spatial location and attribute data portrayed by the GIS are accurate, but can you quantify your confidence in the data? Too often, false confidence has steered the owner and consulting engineer to underestimate the level of work needed to bring the data up to the level for the needed application. Many times, the GIS data may only provide a schematic representation of field conditions and rarely, even for a brief moment, a 100-percent system representation.

The GIS owner should verify the data and record confidence levels. Where did the data originate? Did it come from design drawings, record documents, oral history, or physical hands-on (or really "eyes-on") inspection? Don't just accept the well-enough mentality that will plague the model or design process through to construction.

Prior to taking a stance that the GIS data is "really well," owners need to consider the proposed application for the data and the life cycle costs of maintenance to keep the data in the condition to meet their needs. Then you will have confidence in your GIS.

Larry W. Mitchell, P.E., is a senior project manager with AECOM's South Region in Raleigh, N.C. For nearly two decades he has been managing planning, design, and construction of water and wastewater infrastructure projects for private, municipal, and federal clients.

Upcoming Events

See All Upcoming Events