1987 GIS Spatial Analysis Tools Long

Video Transcript
Download Video
Right-click and save to download

Detailed Description

A 1987 video on available GIS spatial analysis tools


Date Taken:

Length: 00:30:08

Location Taken: Sioux Falls, SD, US


Water – so much a part of daily life that we tend to forget the many ways in which we encounter it and depend on it. People look to water for subsistence, industrial energy, processing materials, low-cost transportation, and for much of their recreation. Life could not exist without it.
In view of projected water demand increases, cities, towns, and rural areas across the nation will need to locate and develop central supplies of potable water. Water pollution may increase under intense agricultural practices, increased urbanization, and industrialization.
The truth is that the available resources of clean, uncontaminated water is dwindling, while demands are rising. A thorough knowledge of hydrologic processes is necessary to evaluate the total water resource, to plan for its management, and to protect it from pollution. 
The U.S. Geological Survey uses advanced computerized geographic information systems – valuable tools for manipulating and analyzing vast amounts of spatial data for solving complex Earth science and natural resource issues. The use of geographic information systems, or GIS, has revolutionary implications for the way the Geological Survey conducts research and presents results.
As the nation’s primary producer of cartographic, geographic, hydrologic, and geologic data, the Survey is using advanced GIS technologies to greatly improve its ability to perform traditional missions of Earth science data collection, research, and information retrieval.
The USGS and other bureaus and offices of the Department of the Interior have established a number of digital spatial databases that are being used in geographic information systems nationwide. These include the National Digital Cartographic Database, Digital Line Graph, Digital Elevation Model, and Land Use and Land Cover Data, the National Coal Resources Data System, the Geographic Names Information System, the Water Data Storage and Retrieval System, and the Rock Analysis Storage System.
The Geological Survey is increasingly being asked to serve as a central repository for digital Earth science data and to be an authority on GIS techniques for arranging, displaying, and analyzing diverse data about such critical issues as the nation’s energy and mineral potential, the assessment of risks from natural hazards, and questions about water supply quantity and quality.
Because GIS technology allows scientists to process and inter-relate many more kinds of data than were previously possible, GIS applications research provides new scientific methods to investigate these issues.
A geographic information system can’t function without digital spatial data. Spatial data consists of the various features that are associated with a geographic Earth location – features that are visually detectable, such as wells, roads, or lakes or boundaries such as county lines or school districts may be represented in the form of points, lines or polygons.
A GIS must be capable of storing and manipulating these types of spatial data. The association of a particular spatial feature with a data type depends on the map or image scale. For instance, a river may be shown as an area at a large scale but merely as a line at a small scale.
Two primary methods, or data structures, are used to organize spatial data within a computer for use by a GIS – the raster data structure and the vector data structure. According to USGS geographic and cartographic research physical scientist Larry Batten, each structure has distinct advantages and disadvantages that affect cost and efficiency.

Raster data would be best viewed as a grid on the Earth’s surface where, say, elevations were sampled, perhaps at a 30-meter grid cell size, so that you would then have a discrete surface of the average elevation within each of those 30-meter cells. 
Vector data structure is sometimes referred to as arc node information. Essentially, it is lines that may either be the lines themselves – for example, roads or streams, or maybe boundaries for physical features such as land use classes, et cetera. That’s maintained as that arc in the computer system rather than, as we talked about, the raster – rather than discrete cells.
There are advantages and disadvantages to raster and vector systems. A raster system is real useful, since you have average elevation, for example, if you can calculate the slopes or the change of elevation between adjacent grid cells. That’s easy to do in a raster system, and it’s fairly difficult to do in a vector system. 
Vectors, though, are better at representing very small changes in a line. If you have a 30-meter grid cell that you’re trying to represent a stream, and your stream actually has a pattern that is maybe only in 5 meters on that, you’ll lose most of that 5-meter meandering, and you’ll only see maybe a straight stream rather than the meanders.
- Each functional component of the principle functions found in geographic information systems has counterparts that work on vector or raster data type. A GIS has yet to be developed that contains all of the possible functions. And it may be argued that no one system should.
According to Batten, the functional components of any given GIS vary considerably depending on its application.
We really have several components within the GIS that’s important. One is data capture, or the way that you ingest information into the GIS. That may come as files – text files, for example. It may come as digital information such as National Mapping Division’s DLG data. It could also come from capturing things using a digitizer, for example, manually tracing the lines that depict boundaries. Once you have the data in, then you need to be able to manipulate and analyze that information.
And there is usually a whole host of functions allowing you to do that. You may also want to be able to display the results using display systems such as this here or out on plotters – to have paper copies are important.
And then being able to update the information that you have in the system is also crucial to most GISs. For example, if land use has changed during the time before it was captured to now, you may want to change that.
Although GIS technology has been in use for more than 10 years, the acceptance level and use of this spatial analysis tool has remained low because of a lack of available detailed base map and digital thematic data. Since data collection is one of the most expensive aspects of GIS technology, the Geological Survey is continuing efforts to merge its existing spatial databases to provide rapid, convenient, and cost-effective access to this tremendous wealth of Earth science natural resources data.
In particular, transportation and hydrologic features digitized from 1- to 100,000-scale base maps serve not only as a fundamental database for GIS studies, but also as a catalyst for the addition of thematic data.
According to the director of the U.S. Geological Survey, Dallas Peck, resource managers across the nation are beginning to recognize the tremendous potential that exists with geographic information systems.
In 1984, the U.S. Geological Survey and the State of Connecticut’s Natural Resources Center agreed to evaluate the use of geographic information systems – GIS technology – to apply Earth science data to solve real-life problems in the state of Connecticut. 
The project was an outgrowth of an ongoing cooperative program between the USGS and the State of Connecticut to gather Earth-related information on geology, soils, water, land use, land cover, and land features. Members of the Connecticut Natural Resources Center combined their expertise with scientists from the U.S. Geological Survey to develop a GIS for this demonstration.
Building on a cartographic base, this involved capturing, storing, updating, analyzing, and displaying computerized natural resources data to solve such problems as well siting and industrial development.
The project ultimately demonstrated how a cooperative approach such as this could be used to address natural resource questions at federal, state, and local levels. Senior environmental analyst with Connecticut’s Natural Resources Center, Sandy Prisloe, described some of the complex environmental questions answered during the course of the project.
We wanted to see how we could use the various types of data layers in a geographic information system to answer questions about the suitability of areas for well siting.
We wanted to see if we could use the geographic information system to assist in development of data for three-dimensional groundwater flow models.
We have some simple applications that looked at how a GIS can calculate estimated seven-day, 10-year low flows for drainage basins in Connecticut. And lastly, we looked at how a geographic information system can do sort of a site analysis and evaluate areas for industrial siting
In order to find the answers to these questions, USGS scientists entered data into the GIS from existing digital databases and from digitizing information from hard-copy maps. Digitization was performed at two locations – the Geological Survey’s Eastern Mapping Center and the Earth Resources Observation Systems – EROS Data Center.
The primary GIS used was the ArcInfo geographic information system.  Sandy Prisloe explains.
We’re very concerned about maintaining the resolution of our source material. We’ve gone to great expense and effort to collect basic natural resource data, and to do it as accurately as possible.
We were interested in using a geographic information system that maintained the accuracy, or the resolution, of the source maps. ArcInfo is the system that does that. We are also interested in using
ArcInfo because it seems to have wide acceptance, at least within the northeast. There are a number of other states that are using the system. So we saw compatibility as a goal there as well.
According to Connecticut environmental analyst Sidney Quarrier, the project area consisted of two 7-1/2-minute quadrangles in north-central Connecticut.
The Broad Brook and Ellington quadrangles make an interesting area for a pilot study because the area is geographically diverse and geologically diverse.
The Central Valley lies adjacent to the eastern uplands. In the valley, a large part of the population lives, and the uplands have a lower population and a more rugged topography. The geology underlying the Central Valley is Mesozoic-age red beds with crystalline metamorphic rocks in the uplands. Groundwater is derived in the valley from thick glacial deposits that were deposited during the retreat of the last glacier.
In the center of the area, these are fine-grain lake deposits and have – supply less water. But along the border areas, large deltas with sand and gravel provide high-yield groundwater aquifers for the area.
Roughly one-third of the database was derived from existing digital information that was available from the Geological Survey. These data include 1- to 24,000-scale digital line graphs, land use and land cover, political boundaries, hydrography, transportation, and state and federal lands.
Sandy Prisloe describes the large amount of information that a GIS must contain to perform the necessary analyses.
We used a number of different data sets for the demonstration project. Some of those existed in digital format already.
Those included things such as a level III land use map that had been put together. Some of the other digital data sets included digital elevation models for the two quads, which is a surface – a rasterized surface data set for land elevation. We also used digital data from National Mapping Division – digital line graph data representing hydrography, the road network, and boundaries. 
Other data sets that were used were the published or map file data sets that we had here at the Natural Resources Center, and they included such things as water table elevations, surficial materials, natural drainage basin boundaries, location of public water supply wells and reservoirs, the boundaries of water utility service areas, location of pollution sources – things of that nature.
The data sets that we had here at the Natural Resources Center were, in turn, digitized by the U.S. Geological Survey as part of the project work.
Four application models were developed to test the usefulness of  this GIS in actual government agency programs.
In the next few minutes, Sandy Prisloe and Larry Batten will demonstrate how these models helped to evaluate the effectiveness of GIS technology in merging data from a variety of sources for use in research, planning, management, and regulatory programs.
The demonstration that we are going to be using today to show GIS capabilities is one that was developed as a part of the Connecticut pilot demonstration project.
The area that we concentrated on was a two-quadrangle area up in north-central Connecticut.
And the map that’s shown on the computer screen shows that area. Other information that’s also displayed on this map is [break in audio] available. And that had been developed through another grant with the USGS and the State of Connecticut. And the various shading patterns that are shown here on the screen and are identified in the key on the side here, those are the actual land uses, or land covers, that were mapped. 
And you’ll notice that there’s a relationship between the shape of this and that little boundary that we saw around the Somers water utility before, well, that boundary around the water utility was actually created by the geographic information system, and it was stored as a map.
That boundary then became the geographic window that we used to look at all the other data sets that we analyzed. So one of the questions that we asked based on land use was, what types of land uses would be unsuitable for groundwater development?
What we were going to do is to go through a series of steps in which we were going to eliminate areas from consideration. We were going to narrow our focus.
And the first thing that we did was to look at the land use and say, what land uses are not compatible with water supply? And we took a very conservative approach. We said if it’s residential use, if it’s industrial or commercial, or if it was agricultural use, we would not consider it.
All right, we were only interested in those areas which are forested or wetlands. And so the geographic information system provided a very nice tool to do that type of analysis. The GIS can, based on the attribute data of what is the land use, can extract and create a new map of just those land uses that are suitable.
And the map that’s on the screen now represents those areas which are either forested or wetlands. And those are the sites that we were going to focus on. This is the first step that we go through. What areas are potentially suitable?
We then wanted to look at some of the cultural information, or environmental information, and how that might affect water supply well siting. And the first thing that we looked at were the locations of pollution sources. 
And the map that’s shown on the screen now represents six known either permitted or accidental pollution events. In some cases, they represent things like salt storage facilities for road salting. It might be an accidental oil spill – something of that nature.
It might even be an industrial lagoon or agricultural new storage facility. And what we did is, we said, for these six sites, let’s be extremely conservative. Let’s pick a buffer – an area around that site of 500 meters. And the map here represents the results of that operation. 
Essentially, the question – or, the operation that we used the geographic information system for was to develop these buffers – these areas. And these buffers – the system creates these buffers just like it created the buffer around the entire water utility. Once we’ve created the buffers, we can essentially go through an operation which is analogous to an erase. Take the buffers, overlay those on top of the areas that are suitable based on the land use characteristics, and erase areas that were suitable but which fell inside the buffered areas, and eliminate those from consideration. And the results of that are then shown quite graphically here.
There were a number of red circles that went down the middle of the study area, and those are gone now. Okay, we also – as well as pollution sources – point sources, we also looked at the water quality of streams in the area. The area – the streams shown in blue here are also – have a buffer around them of about 100 meters in size. And those are streams that are waste-receiving streams based on the water quality classification program in Connecticut.
Most major streams are classified as an A-, a B-, or a C-quality stream. A B- or a C-quality stream means that it is receiving a waste discharge – either a permitted waste discharge or a leachate from a landfill – something of that nature.
We created a 100-meter buffer around those streams based on their classification – if they were class B or class C. All right, again, we went to the attribute information that defined something about this linear feature, being a stream, and said, if you are a class B or a class C stream, create a 100-meter buffer around it.
And, as we did with the pollution sources, we then subtracted this area from consideration. We did comparable things for municipal zoning. If it was zoned for industrial use, we wouldn’t consider it. If it was zoned for commercial use, it was eliminated. We also looked at where public water supply wells were currently located and created buffers around those wells and said, we’re not going to put another well adjacent to an existing well.
So we went through a number of steps in which we kept narrowing the focus, narrowing the area that would suitable, all based primarily on what was going on the land’s surface. Then what we decided to do is, now that we’ve eliminated everything, we’ve got to find out what’s good. And so we looked at some of the hydrogeology. And the map that’s on the screen represents the surficial materials for the area. And what we were interested in looking at for the surficial materials were areas that had primarily coarse-grained stratified drift deposits – areas that would have large volumes of water contained in those deposits.
We were not interested in tills, which have poor water-bearing capability. We were also not interested in any fine deposits. So we were looking for primarily coarse-grained saturated materials. And to find out where those are, we start with a detailed surficial material map.
From this, we were then able to extract those areas which were just coarse-grained materials. We also were interested in knowing not only what is the texture of the material, but to what depth is it saturated?
And, Larry, you’d probably do better at explaining how this particular map was generated since you were intimately involved in producing it.
This map is a derived map product. By derived, I mean we’ve taken several different layers of data, combined them together, to give us a new product here. What we have is taken digital elevation model data from the National Mapping Division of the USGS to give us an idea of the surface topography. We’ve then compared that, in a raster processing system, to the bedrock and water table elevations, subtracted those together to find the saturated thickness or the column of water-bearing material above the bedrock.
We’ve then brought that raster data into our vector system where we’re doing the analysis here, and dropped those boundaries between these grid cells that were within the same contour interval. Here we’ve broken them into 20-foot contours. And that’s what the colors here depict. What I’d like to do is zoom in here to show you how a rasterized vector data would look like.
What we have is the result of the original 30-meter grid cells that have given us this rough shape. Most of the vector data we’ve seen would give us a very smooth edge to that.
One of the advantages of – or, one of the things that we were very interested in Connecticut when we were looking at geographic information systems to start with is some of the differences between the raster and vector data. I think a point to keep in mind is that, with vector data, you can go to a raster format. You can’t go from raster to vector and maintain the resolution of your original data.
That’s right.
So the nice advantage of the vector-based systems is that you can maintain the resolution of the source material you’re starting out with. Anyway, what we – what we had to do at this point was to combine the thickness of the saturated materials with those areas which were coarse- grained and come up with essentially all those areas which were coarse-grained stratified drift deposits with greater than 40 feet of saturated thickness.
And that was – by looking just at hydrogeology, we had narrowed down now the scope of where potential sites would be based on, was there water there, was it coarse-grained material, whatever. We then had to take the results of that and combine it with those areas which were suitable based on the land characteristics.
And so the resultant map of all these various steps that we went through – this one that’s being drawn on the map – on the screen now. The areas that are highlighted in blue – there are six of them – they’re all areas that met all of the decision criteria that we initially started out with.
We’ve looked at land use characteristics, the location of pollution sources. We looked at the hydrogeology of the area as well as some other things. And by going through this series of steps, we’ve really narrowed the focus now to six potential sites. This information is also being displayed, at this time for reference purposes, with some other of the basic data layers that we started out with. And they include the digital line graph hydrography and the digital line graph road network. That’s displayed in red.
And also, we’ve superimposed on this the location, or the names, of features that are shown on the topographic maps. And those are contained in the Geographic Names Information System. So what we’ve done, essentially, is put together a composite map now – a map that represents all those areas which are potential sites for groundwater development
And we’ve combined that with basic information from the digital line graphs. And we have pretty much a complete composite map that’s of adequate detail that we could go out and do a reconnaissance-level survey of, where are these sites in the field – do some double-checking to make sure that what we’ve come up with is actually what’s out there. All right.
What I’ll do now is zoom in on this area again to show you the actual detail around each of these sites. For example, we can go in this region here. And you can see that, were this a site we were interested in, we’d be able to know exactly the location and the corner we’d want to turn onto in this road network.
Another important feature to bring out within this geographic information system type of analysis is that, if you came down with this result, and for some reason, learned that this was not a good location –perhaps there was a rare and endangered species or something that was found within that area.
You could then go back, very easily, and change some of the criteria that you had previously used and change the outcome here. 
That’s something that would be easy to do in a GIS but much more difficult to do in a manual mode.
It would also be possible – and sort of following along the same lines, as new information becomes available, that that can also be incorporated into the process. And I’d go back to the point you raised about rare and endangered species.
If that information were mapped, and it became available, that could be thrown up – or, combined with this almost – well, not instantaneously, but very quickly.
And then the planner, who is going to be the one that actually uses this information down the road, will be able to factor that information in as well.
I think that one thing that’s worth noting with this is that the application that we’ve demonstrated here was for one small water utility located up in the top corner of the Ellington quadrangle.
The process that we went through – the various steps that we used – those are all repeatable.
And we could do the same analysis – not just for that one water utility, but for all of the water utilities that were displayed on those two quadrangles.
Or, better yet, if you’re covering a larger geographic area such as a public water supply management area in the state of Connecticut, you could conduct this type of analysis for each of the utilities – and there could be 40 or 50 of them – in a much larger area. The steps would be the same.
The process would be the same. It would take a little bit longer because of the computer resources you’d be using, but the planner would then have results, not just for one water utility, but for a large area.
And I think that’s one of the tremendous advantages that this technology offers. 
Geographic information systems applications are numerous and diverse. Federal agencies use these spatial analysis tools to manage national parks, national forests, wildlife refuges, and federal mineral and petroleum resources.
In addition to Connecticut, other states and local governments use geographic information systems to plan industrial development and recreational activities. Private organizations find them useful to target potential market areas and to determine site suitability for commercial development.
If development and management of our water and other finite natural resources are to be based on the best available information, it’s crucial that this knowledge be transferred to policymakers in a timely, understandable, and cost-effective manner. Geographic information systems provide a powerful tool for collecting, manipulating, displaying, and analyzing large volumes of spatially referenced data.
The U.S. Geological Survey’s Geologic, Water Resources, Information Systems, and National Mapping divisions continue to combine their expertise to lead in developing techniques for applying advanced GIS technology to solve resource problems encountered by government, private industry, and the public.