USGS Multimedia Gallery: (Video)--Overview of the National Hydrography Dataset and The National Map

Hello, my name is Jeff Simley with the U.S. Geological

Survey. We are going to talk about the National

Hydrography Dataset and The National Map. So what is

The National Map? Well, let us think of it this way.

Let us think of it as a tree and The National Map is

the trunk of the tree; it is kind of the mother board

for a lot of information, a lot of geographic

information. Information such as elevation data,

structures data, transportation information,

orthoimagery, and hydrography. So these different

types of geographic information fit into this platform

that we call The National Map. Kind of like the

branches connected with the trunk of a tree. In

hydrography we have the National Hydrography Dataset

and the Watershed Boundary Dataset components that

make up the hydrography theme of The National Map. In

the National Hydrography Dataset we have things like

lakes and streams; and connected to lakes for example

might be information on the water chemistry of the

lake. Connected with streams we might have information

on fish habitat in the streams. An example might be

say cutthroat trout in a stream. So we need to access

this information and this can be done through

something like a map portal. So a map portal allows

us to enter in this geographic information system and

to access information such as where the cutthroat

trout are in the United States. We do this through The

National Map which adds a component, it has

hydrography and the National Hydrography Dataset. So

you can think of the National Hydrography Dataset very

similar to the GPS you have in your car. In the GPS in

your car, there is a 4 million mile road network that

is connected to that GPS and it allows you to solve

problems. Say for example we want to find out where a

gas station is. We are at a certain location and we

want to find the route to a gas station. This is done

by using this 4 million mile road network. We see for

example here a bunch of streets that make up this

network. The GPS knows where our position is on the

surface of the earth but what we really want to find

out is where we are on the network of roads. Not just

where we are in space but where we are on a network.

And so it snaps to the network and it finds our

network address on the network, so now we know where

we are on the network. The next thing it wants to find

out is where are the gas stations. The gas stations

are also located on the network. Then it tries to

solve problems of how to get from where I am now to

the nearest gas station. There are a couple of

solutions here and it picks the optimal solution and

then plots that solution for us and solves a problem,

it tells us how to get from where we are to a gas

station. The streets that make up this GPS system

really makes all of this possible, and then the

linking of information to those streets and knowing

where I am on those streets is what makes this work.

We do the same thing in hydrography but instead of

using a 4 million mile road network we use a 7.5

million mile stream network. So we can access this

information through a desktop portal such as a GIS

system or something like StreamStats as opposed to the

GPS system that you have on the dashboard of your car.

This is an example of The National Map viewer, another

way of accessing this type of information. Someday we

will be able to put this in the hands of people in the

field on mobile devices but we are not quite there

yet. We are still tied to a desktop application. So

this is basically what the NHD is all about, it is

things like lakes and streams. As you can see in this

map here there is this large lake and a big stream

network around that lake. You can see that there is a

ditch, a diversion tunnel, there is a marsh, there are

stream gages, and dams. It is the kind of information

that you would find on a USGS topographic map for

example. Basic features that make up the surface

water hydrography of the United States. The stream

network is set up into a vast network of stream

connections. It is organized into a hydrologic unit

which you see here. This hydrologic unit is

surrounded by a ridge that more or less kind of

divides the water and has all the water within this

drainage area flowing out in one direction. So most

of the hydrologic units of the United States looks

like this; they have water flowing out in one

direction. That ridge that we see there forms what we

call a hydrologic unit and that hydrologic unit is the

foundation for the Watershed Boundary Dataset. So this

is an example of that unit and the surrounding units

that are drainage areas. This particular unit is

called a Subbasin or a HUC 8 which is the 4th level in

the hierarchy of streams. We can further subdivide

this Subbasin or this hydrologic unit into more

detailed units. And then we can further divide these

units into even more detailed units. So if we look at

the hydrography and the hydrologic units on top of

each other we can see how they are integrated

together. You can see how there are individual stream

networks that drain within these hydrologic units. So

this is an example for the hydrologic units of the

United States at the 4th level of the hierarchy. There

are 2,246 of these over the United States. Here we are

just looking at the conterminous United States, but

the National Hydrography Dataset is one nation-wide

seamless dasaset. So although it is composed of all of

these units put together, it is one continuous

seamless dataset for the country. This is an example

of what the NHD looks like. It is basically streams

and we know some information about these streams. For

example, we know that this is Swan Creek, we know that

some of the streams are perennial, some are

intermittent and some are ephemeral. We also know the

flow direction of all these streams. So this entire

7.5 million mile flow network for the United States,

we know the flow direction, the direction of flow of

the water on all these streams. So for example we can

take a look at a spot, this green square you see there

at that location we can easily identify all the

streams upstream; those are those red lines. We can

also go downstream and trace the pathway downstream

from this green dot. Say for example we had a toxic

spill at this location we could trace the water

downstream and see what it would affect downstream.

One of the problems we have in creating this dataset

is to try and figure out how water goes though lakes.

So if you look at this map here we need to figure out

how water is entering the lake and where it is exiting

the lake. So we use a system of artificial paths that

give us the flow direction of water routing through

the lake so we can get a complete network, even

through all these polygons that make up the lakes of

the United States. One of the things we do is we

identify every stream segment with an identifier

called a ReachCode. This is a 14-digit number. The

first 8-digits tell us the hydrologic unit that we are

in. So in this case here hydrologic unit 14, the first

2-digits is the Upper Colorado River, hydrologic 01 is

the Colorado Headwaters, and then the 00 after that

is not used, and then the 02 is the Blue River. So it

is the Blue River of the Colorado Headwaters of the

Upper Colorado River Region. Every stream segment of

the United States has an identifier like this. So even

if the stream does not have a name we still have a way

of identifying that stream. Something else we do is

we take each one of these identifiers, these

ReachCodes, and we divide it up into an address range

with 0 at the downstream end and 100 at the upstream

end, then we equally divide it up into units between

0 and 100 no matter how short or how long or how

sinuous the stream segment is, it is always a 0 to

100 address range. You can take any point along that

stream and give it a value as to its address range on

the stream. So as an example we have a USGS stream

gage. We know where it is in space, we know the

latitude and longitude of that stream gage. What we

really want to do is to snap it to the network. So

much like the GPS in your car we know where you are

but we really want to know where you are on the

network. So we snap the stream gage to the network and

we get the network address. So in this case here, the

location of that stream gage is on reach

14010002000421 and its measure upstream, its location

upstream, is 19.8392% upstream on that stream segment.

So now we know the address of the stream gage on the

nations rivers network. There is only one spot in the

United States that has this address. It is very

simple. Two identifiers can tell us exactly where

anything is on the network. So for example here is a

stream gage on the network. The stream gage makes the

stream network more intelligent because the stream

gage can tell us how much water is in this river, and

likewise the network can make the stream gage more

intelligent. So we know what is downstream, what is

upstream, we know the name of the stream that we are

on. We know that there is a dam upstream. We know that

the dam is impounding Green Mountain Reservoir so

there is a reservoir upstream of this stream gage. We

know how many acre-feet are in that reservoir. Further

upstream we know that there is another reservoir that

has a diversion that is diverting 300 cubic-feet per

second of water out of the stream system. We know how

many miles of stream are upstream, we know how many

miles of perennial stream, how many miles of

intermittent stream, how many miles of ephemeral

stream. We know the drainage area, we know that there

is a heavy metals tailings pond upstream of this

location. So all of this information on the network

makes the stream gage more intelligent and the stream

gage makes the river more intelligent by telling us

how much water is in the stream. This is an example

of a hydrograph from that stream gage telling us the

stream flow for that river. We also want to talk about

how this data works in an information system. So in

the GPS in your car, for example, we have data such

as your position, where the gas stations are,

where the roads are. We can transform that data into

information that gives us more intelligence, so not

just my position but my position on the network. We

can then further transform that information into

knowledge, so the geospatial processing system that is

on the dashboard of your car takes this information

and process it to give us knowledge such as how to get

to a gas station. Using the National Hydrography

Dataset and other types of data we can do the same

type of thing in a geographic information system. We

can take information on land cover, temperature,

precipitation, elevation data, hydrography data such

as the NHD and process this raw data into information,

and then we can take this information and use a

processing engine such as StreamStats to determine

knowledge such as the discharge of water at an ungaged

site. This is an example of StreamStats, it is a very

advanced systems that is interactively operated over

the internet, that allows you to pick any point along

a river and determine the stream flow at that

location. On the left there you can see that we have

picked a location and that pink shaded area is the

drainage area that drains through the point that we

picked the little red star you see there. Then on the

right we can see information on stream flow. At the

very bottom of those tables at the right you can see

predicted stream flow at that location for the 100 yr

flood interval would be 3,680 cubic feet per second.

And we can do this by using different types of

geographic information combined with hydrologic

modeling. A typical problem that we want to solve is

how do points A, B, and C relate to each other. Point

B is a drinking water intake, point A is an

industrial discharge and point C is a pesticide

application. And on the left there you can see that we

have this model of the NHD/WBD database. We can take

water observation A and realize that it is upstream of

water observation B by the network that is within the

flowline of the NHD dataset. Point C is a pesticide

application; it is not directly on a stream but it is

on a hydrologic unit and we can relate the hydrologic

unit to the flowlines and understand that C is

upstream of B. So we know that A and C affect the

drinking water intake B. That is basically the types

of problems we are trying to solve.