In this lab, you will continue to practice downloading, manipulating, mapping, and analyzing hydrology data that is publicly available online to continue your study of the Buffalo-San Jacinto watershed subbasin. Specifically, you will work with elevation data from the National Elevation Dataset (NED) accessed through The National Map Viewer and rainfall data from the National Climatic Data Center (NCDC) accessed through Climate Data Online (CDO). You will also learn how to calculate statistics by watershed, such as mean elevation and mean annual precipitation.
You will begin by creating a new map for Lab 2.
Before downloading any new data, you will further process data from Lab 1 in preparation for this lab.
Notice that the layer is in a geographic coordinate system called NAD 1983, which stands for North American Datum 1983. Because the data has a geographic coordinate system, the coordinates are stored in degrees, which indicate the three-dimensional location of the data on Earth's spheroid. Though the data itself is stored in a geographic coordinate system, your computer monitor is flat, so, even though no projection has been defined, the data must be displayed in a particular projection. Whenever ArcGIS displays data in a geographic coordinate system, it uses a pseudo plate carrée projection, where one degree of latitude by one degree of longitude is represented as a square, rather than a curved trapezoid. In other words, all lines of latitude and longitude are evenly spaced. This type of projection results in stretching in the east-west direction, which increases the farther north or south from the equator you are mapping.
Working with geographic coordinate systems is fine for creating purely visual maps, as you did in Lab 1 (though the visual distortion can be disorienting and misleading), but, in this lab, you will be calculating areas, distances, and overlaps between features. Such calculations require the three-dimensional coordinates to be projected down onto a two-dimensional plane, so that the coordinates are stored in linear units, such as feet or meters, rather than degrees. In order to facilitate measurements of distance and area, you will now project the Watersheds layer into the State Plane Texas South Central projection which is best suited to mapping the greater Houston region.
Because both the input and output coordinate systems are based on the NAD 1983 geographic coordinate system, no geographic transformation is required.
Now that you have the correctly projected layer, you no longer need the original NAD 1983 layer.
Notice that the layer is now in a projected coordinate system, NAD 1983 StatePlane Texas S Central FIPS 4204 (US Feet).
You may have noticed that the visual appearance of the watersheds did not change in your Map Display, even though you projected them. That is because the data frame takes on the projection of the first layer added to it. Since you first added the original unprojected Watersheds layer into the data frame, which was in NAD 1983, the data frame still displays the data in NAD 1983 (or psuedo plate carrée). Currently, the projected Watersheds_StatePlane layer is being projected-on-the-fly back into NAD 1983 for visual purposes.
Move the cursor around the screen and notice that the coordinates at the bottom of the map view are shown in decimal degrees. This is another clue that the data frame is still using a geographic coordinate system; however, you would like the data frame to display data using the local Houston projection.
At the top of the Contents pane, double-click the Lab 2 map to open the 'Map Properties' window.
While you could search for or navigate to the State Plane Texas South Central projection, as you did before, in this case, you know that the same coordinate system is already used by the Watersheds_StatePlane layer. In such an instance, it is often easier to import the coordinate system from another known layer, especially if you are not familiar with the hierarchy of the coordinate system folders
Scroll to the top of the 'XY Coordinate Systems Available' list.
Expand Layers.
Notice that all of the coordinate systems used by layers currently in the map are displayed. You can expand the coordinate systems to determine exactly which layers are stored in which coordinate systems.
Click NAD 1983 StatePlane Texas S Central FIPS 4204 (US Feet).
Notice that the watershed boundaries are now more compact in the east-west direction, as expected, because the local projection results in less distortion than the pseudo plate carrée projection used to represent geographic coordinate systems.
Now you are ready to download digital elevation model (DEM) data for the Buffalo-San Jacinto subbasin. Some local government agencies, such as the Houston-Galveston Area Council (H-GAC) contract to have LiDAR data collected, which provides high resolution data containing both the elevation of the bare land and the heights of features in the built environment. Though we are lucky to have this high-quality data available in this particular region, for projects anywhere in the U.S., the best available DEM data generally comes from the National Elevation Dataset (NED) produced by the USGS. More information regarding NED data can be found at ned.usgs.gov. You will download NED data from The National Map Download Client.
First, you will constrain your downloads to your area of interest.
After zooming into Houston, you can see the blue boundaries of the individual subbasins and each subbasin is labeled with its HUC-8 number on the map.
Your data search will now be constrained to the polygon for subbasin 12040104.
Next, you will select the data products you are interested in viewing and downloading.
You are now provided with a listing of all the DEM tiles covering the area of subbasin 12040104. Although 12 results are listed, there are only 3 areas of coverage provided in 4 different file formats: ArcGrid, GeoTIFF, GridFloat, and IMG. You will download only the 3 files in the ArcGrid format
Since you just added new files to your folder, you will need to refresh it in order for them to appear in the Catalog pane.
Each raster file with a name such as grdn30w095_13 corresponds to a 1x1 degree tile. The file name contains “grd” for grid, followed by the latitude and longitude of the top left corner of the tile.
Adding raster data in ArcMap
You will be asked if you would like to create pyramids. Pyramids cache the raster at multiple reduced resolutions, resulting in an increased file size, but better rendering performance in your map view. It is normally a good idea to create pyramids, but since you will be processing the corresponding rasters into a new file in the next step, you will not create pyramids at this time.
Notice that the 1x1 degree tiles appear as angled rectangles, because they are being displayed in the State Plane Texas South Central projection. If the data frame were still in the NAD 1983 geographic coordinate system, the tiles would look like squares. Now you will look up the native coordinate system of the raster files.
Notice the Geographic Coordinate System is NAD 1983 and that no projected coordinate system is listed. Since this coordinate system is not the same as the one currently used by the data frame, the raster layer is being projected-on-the-fly into the State Plane Texas South Central projection.
The edges where the three tiles meet are currently visible, because the minimum and maximum elevation is different in each raster, causing the same value to be represented with a different shade of gray in each raster. To solve that visual problem and also simplify future processing steps, you will mosaic the three rasters into a single raster. Before creating a mosaic, you will need to look up some information from the original rasters.
Notice that there is 1 band, meaning that each pixel only stores a single value: the height above sea level in meters. Aerial imagery has 3 bands to store the 3 RGB values. The format is listed as GRID, which stands for an Esri Grid, which is a raster file format native to Esri software. The pixel type and depth is 32-bit floating point, which indicates that the cells can store decimal data.
The mosaic may take a couple minutes to process. When it is complete, notice there are no longer visual seams between the tiles in the mosaicked raster. Now that you have a single mosaic, you no longer need the three originals tiles.
Now you need to project your mosaic into the State Plane Texas South Central projection, so that you can properly calculate spatial statistics based on the data it contains.
You should now see the NAD_1983_StatePlane_Texas_South_Central_FIPS_4204_Feet displayed for the Output Coordinate System.
For cell size, you may normally want to keep the original resolution, which was approximately 30 m, but, in this case, you will reduce the resolution to expedite processing times during this lab. The cell size is always specified in the same units as the projection, which in this case is feet, so you will select a cell size of 300 feet.
Remember that the data frame was already displaying all layers in State Plane Texas South Central, so you should not notice much of a difference between the two layers, other than that the cell size has increased, meaning the resolution has decreased.
Now you are ready to clip the DEM mosaic to the Buffalo-San Jacinto subbasin.
Checking that box ensures that the raster is limited to the actual shape of the watersheds, rather than a rectangle covering the same extent.
The DEMSubbasin layer is now clipped to the shape of the Buffalo-San Jacinto subbasin.
Right now, the elevations in the DEM are in units of meters. In order to convert these units into feet, you will use the raster calculator.
Visually, the meters and feet layers should be identical, but, if you look at the layers in the Contents pane, you will notice that the original layer goes from elevations of -8 to 62 meters and the newly calculated layer goes from elevations of -26 to 204 feet.
Now you will determine which portions of land may be affected by a 15’ storm surge. Obviously, complex inundation models will take more variables into account, but, in this instance, you will simply highlight all the areas of land with an elevation of 15’ or less.
Now the cells containing an elevation of 15 feet or less are highlighted on top of the elevation raster.
FOR MAP LAYOUT TO BE TURNED IN
Create an 8.5 x 11 layout showing the DEM clipped to the subbasin with areas less than or equal to 15 feet in elevation highlighted.
Now you will create a new map document from the one you are currently using.
Patterns within the data are now easier to see, especially at the lower elevations.
In addition to raster DEM data, it is sometimes useful to be able to represent elevation using vector contour lines. Contour lines at any regular intervals or discrete values can be created from DEM data.
The base contour is the lowest contour that will be shown. Since the lowest elevation is -26 ft, you will set your base contour to -20 ft.
Since the x,y coordinates are in the units of your State Plane Texas South Central projection, which is feet, and you have also used the Raster Calculator to convert the z units stored within the raster cells to feet, you do not need a custom Z factor.
Now you will create a hillshade raster, which provides a shaded relief of the terrain based on a certain sun angle. It stores a value between 0 and 255 indicating the extent to which the cell would be shaded from the sun.
The azimuth and altitude refer to sun angles. For now, you will stick with the default values.
As previously mentioned, the Z factor is used when the horizontal units of the projection and units of the elevation measurements stored in the raster cells are not the same. In this case, both are measured in feet, so a value of 1 is technically correct. Typically, hillshades are used to create realistic three-dimensional representations of mountains, which can be much easier for an audience to interpret than a flat DEM or contour lines. Unfortunately, the opposite is true in Houston, because the area is so flat. It would be difficult to see any changes in elevation using a hillshade, since few shadows would be cast by the terrain. To compensate visually for the flat terrain, you will exaggerate the vertical elevations.
Typically hillshades are shown beneath transparent layers conveying other information, just to give the map a realistic appearance.
Now the gradual changes in elevation are apparent from the coloring and are complimented by a realistic display of the terrain created with the hillshade.
As a final touch, you will add the flowlines you downloaded in Lab1 to your Map Display.
FOR MAP LAYOUT TO BE TURNED IN
Create an 8.5 x 11 layout showing transparent elevation in graduated colors on top of a hillshade, with watershed boundaries and flowlines visible.
Now you will calculate elevation statistics by watershed.
This table tells you the statistics regarding all the elevation values within each watershed zone.
If you save this table inside your file geodatabase, you will not be able to open it outside of ArcGIS, so you will instead save it as a text file outside of your file geodatabase.
Now you will open the exported text file in Excel.
Continue formatting the table until you are satisfied with its appearance.
FOR TABLE TO BE TURNED IN
Create a table highlighting the minimum and maximum values for the minimum, maximum, range, mean, and standard deviation of all elevation values within each watershed.
Now you will download rain gauge station data created by the National Climatic Data Center (NCDC) using the Climate Data Online (CDO) interface.
You will search for data using the NHD hydrologic units. Previously, you had been working with the Buffalo-San Jacinto subbasin (HUC = 12040104). In this case, you will step up two levels to the Galveston Bay-San Jacinto subregion (HUC = 1204).
Check your email. You should receive two emails a couple minutes apart, although it may take a few hours to receive the second email. The first one indicates that your data request was submitted and the second one includes the requested data.
The first column contains the unique station identification code and the second column contains the station name. Next are the elevation, latitude, and longitude of the stations. The annual precipitation field contains long-term averages of annual precipitation totals in hundredths of inches. More information is available on the Data Set Documentation and Samples portion of the CDO website. The annual precipitation field "ANN-PRCP-NORMAL" may not be in the F column, it might be in the BP column, in which case you will need to move it there using the copy-paste function.
Before opening this table in ArcGIS, you must reformat some of the field names, which cannot have special characters and must be 13 characters or less.
Now you are ready to start a new map document and display the tabular rain gage data you just downloaded.
Because the coordinates are in the form of latitude and longitude in decimal degrees, you know you will need to select a geographic coordinate system, rather than a projected coordinate system. While the data could theoretically be in any geographic coordinate system, you will select the North American Datum 1983, commonly abbreviated NAD 83, because this is coordinate system of the data provided on the NCDC website.
The points should now appear on top of the watersheds, though they also extend beyond the watersheds in the Buffalo-San Jacinto subbasin, since we downloaded them for the entire Galveston Bay-San Jacinto subregion.
Since the points appear to be in reasonable locations (rather than in another country or the middle of the ocean), you will want to export them to a new feature class in your ElevationRainfall geodatabase. Exporting to a feature class will allow you to reuse this points layer in other future map documents without having to go through the display XY data process each time.
Because you last exported a text file outside of your geodatabase, notice the output feature class is now set to a shapefile in your Lab 2 folder, rather than a feature class within your ElevationRainfall geodatabase.
Since you are now using a permanent feature class, you may remove your temporary Events layer and the corresponding Excel table.
Repeating the technique you learned earlier in this lab to project vector data, project the PrecipStations layer into the State Plane Texas South Central projection. Save the resulting feature class and name it “PrecipStations_StatePlane”. Remove the original PrecipStations layer from the Contents pane.
Now you will calculate the mean annual precipitation over each watershed using Thiessen polygons, which associate every cell in the watershed with the nearest rain gage.
Before populating the variables in this tool, you will change an Environment setting so that the polygons are calculated for the entire region that you just zoomed to.
You will notice that polygons now fill the entire Map Display indicating which areas are closest to which rain gages.
Notice that all of the fields that you originally downloaded from CDO are still included, because you selected to output all fields when running the Create Thiessen Polygons tool. If you do not see all of the same fields, re-run the tool and this time output all fields.
In order to determine which portions of the resulting polygons overlap with which watersheds, you will now perform an intersect operation between the two layers. The result will allow you to calculate weighted averages of the precipitation in each watershed.
The resulting layer integrates all of the boundaries from both the Thiessen polygons and the watersheds, limited to the extent of their overlap.
Ensure your ‘Field Calculator’ appears as shown below and click Run.
Right-click the HU_10_NAME field name and select Summarize….
For statistics fields, use the drop-down menu to select the Shape_Area field and select Sum.
The resulting table gives the numerator and denominator in the equation for each watershed.
Repeating the techniques you just learned, add a new field to the WatershedPrecip table called “Precip” of type double. Use the field calculator to evaluate [Sum_APProd]/[Sum_Shape_Area]. The result is the precipitation for each subwatershed. Export the table to Excel and format it. Also include the mean annual precipitation over the entire watershed.
FOR TABLE TO BE TURNED IN
Create a table containing the weighted mean annual precipitation for each watershed, as calculated using Thiessen polygons, along with the total mean annual precipitation over the entire subbasin.
Now you will calculate precipitation for each watershed using a different interpolation method.
The result is a solid surface estimating the rainfall at each cell, based on the data collected at each rain gage. Turn the Thiessen polygons back on to give them a hollow fill. Symbolize the rain gages as you desire.
FOR MAP LAYOUT TO BE TURNED IN
Create an 8.5 x 11 layout showing the Thiessen polygon boundaries, interpolated rainfall layer and rain gage locations.
TOP MAP WAS THE ORIGINAL MAP, BOTTOM IS MY SCREENSHOT WHICH HAS SOMEWHAT DIFFERENT DATA. NOT SURE IF IT MATTERS BUT THOUGHT IT WORTHWHILE TO AT LEAST CHECK
Using the techniques you learned earlier in this lab, use the "Zonal Statistics as Table" tool to export a table containing the total annual rainfall for each watershed, as calculated from the AnnPrecip spline interpolation layer and format the table in Excel.
FOR TABLE TO BE TURNED IN
Create a table containing the total annual precipitation for each watershed, as calculated using the spline interpolation method, along with the total annual precipitation over the entire subbasin.