The highlight of yesterday’s Geospatial Summit for me was mention of the National Hydrography Data Set.

Tommy Dewald (U.S. Environmental Protection Agency) and Keven Roth (U.S. Geological Survey, retired) set about the task of developing a surface water dataset of the nation in the 1990s with a vision to create a solution for the 21st century, when suitable water resources would become critical. What oil was for the 20th century, water would be for the 21st century.

They methodically canvassed water scientists across the country to determine the elements that would create the solution they had envisioned. By 2007 this effort had been completed with over 25 million features in the data set covering 7.5 million miles of streams and 6.5 million lakes.

By 2007 the number of scientific applications taking advantage of the unique analytical powers of the NHD had grown to cover all aspects of hydrology, pollution control, resource management, and fisheries biology.

In the next four years the NHD had become even more widespread throughout the sciences to the point where any serious study of water resources demanded the use of the National Hydrography Dataset.

The result of their nearly 20 years of work has become a truly national hydrography dataset for which they received the 2011 USGS Henry Gannett Award given annually to recognize especially distinguished contributions to the topographic mapping of the Nation.

So what are we seeing in this graphic and why is it so important to us?

We are all familiar with a roadmap showing major highways on down to local roads and even some things that are barely roads. Well this graphic is a map of nearly all the streams, lakes, and small tributaries in the continental United States.

In addition, what you do not see is the data behind this lines and shapes which captures water flow information that is needed for modeling applications like mentioned above. So imagine that you just can’t draw a line on this map but you have to have data for it as well and you begin to grasp the scope of this dataset and the effort to build it!

I tried reviewing that data by downloading the NHD file, unzipping the Shape file (a spatial data file format), and importing it into Spotfire and using its map table and map functions to display it. The Spotfire file is 244 MB and is an an example of “big data” in memory on the Web.

You can see the initial graphic and more information about the NHD’s and EPA’s work here.