CE 547 - GIS in Water Resources Engineering

CE 547 –GIS

due April 1, 2009

Assignment #6

Working with Statsgo Soils Data and Land Use

This exercise is intended to introduce you to soils data from the STATSGO soils coverage as well as land use data. In addition, you will learn about joining and relating tables and about some geoprocessing commands.

 For this assignment you will again use the headwaters of the Pecos basin.  You’ll use the STATSGO coverage for New Mexico.  You’ll “clip” (using the ArcToolbox) a portion of STATSGO to conform to the watershed boundaries of pecoshuc (which you worked with in the previous assignment). Tabular data on the soil properties are provided by the Info tables Mapunit, Comp, and Layer, which are held separately from the soil coverages in the Info directory. Finally, the exercise looks at the land use in the study region and the coinciding characteristics between land use and soil type.

Procedure for the Assignment

1. Getting the Data

You may download the files from the CEGIS account.

Alternatively, download nm.tar.Z from the STATSGO website

This website is linked from the class links page.  Under 'Data', select 'ArcInfo coverages'.  The coverages are zipped and tarred.  To unzip the file, you need to use WinZip or a program that can untar the files (WinZip can be downloaded free).  The zip that comes with the Windows XP operating system cannot untar.  Select all and extract to the directory where you plan to do your work.

 Viewing the Data

 Start ArcMap.  Click the Add Data button.  You will need to connect to the folder where you extracted the STATSGO coverage.  Note the directory structure of the coverage.  You will need to be in the folder named ‘spatial’ to properly access the data.

 The New Mexico STATSGO coverage has 2084 polygons.

 Clearly this is an immense data base, but the large size of its polygons makes it suitable for analysis of fairly large areas which contain several STATSGO polygons. There is another database called SSURGO, still under development by US Dept of Agriculture, which details soil polygons at the component level rather than at the map unit level.  SSURGO currently has limited coverage in New Mexico.   You can find out more about SSURGO data on the NRCS website (http://soils.usda.gov/survey/geography/ssurgo/) .

 
Change the symbology so that your data frame has more meaning.  Next, you need to add the headwaters of the pecos watershed.  (you may have called this pecoshus in the last assignment).  Change the symbology of pecoshuc so that the polygon is hollow.  Note that some of the polygons of STATSGO correspond with watershed boundaries (i.e. rock outcroppings along edges, polygon following the river valley, etc).

 Next, we’ll use the ArcToolbox to clip the boundaries of STATSGO to that of our watershed.  ArcToolbox > Analysis Tools > Extract > Clip.

Follow the dialogue boxes as shown below.  Click the ‘Show Help’ to learn more.

 Now, you’ll have 86 records in your table instead of 2084.

 Next, we’ll add the tables you need to really get the information that STATSGO offers.  Once again, the Add Data button is used to add tables to a map document.  Use the control key to select three tables at a time:  comp, layer, and mapunit.  You may want to go back to the STATSGO website and view the Users Guide to find more about the tables and the information available in each table.

Once the tables have been added, the Table of Contents window will automatically switch to the Source tab as shown below.  Take a look at all of the tables to get an idea of the information they contain.  (Right click for the context menu which will allow you to open the table.)  The attribute table for your clipped STATSGO layer is of limited use without the other tables.  So our next step will be to provide the appropriate joins and relates to get the information that we want.

 
Some of the fields in the comp table are as follows:

     ·        Muid - the mapunit id number e.g. NM963

·        Muname - the descriptive name of the map unit

·        Seqnum - a number from 1 to 21 that identifies which soil component of this map unit is being described.

·        Compname - the name of this soil component.

·        Comppct - the percentage of the map unit occupied by this component.

·        Slopel - lower limit of surface slopes for this component

·        Slopeh - upper limit of surface slopes for this component

·        Surftex - USDA soil texture type (C = clay, S = sand, Si = Silt, L = loam, F = fine, G = gravel, and combinations of these include CL for clay loam, SL for Sandy Loam, etc)

·        Hydrgrp - hydrologic soil group (A = sandy, free draining soil, D = clayey, poorly drained soils, B and C are intermediate soil groups.)

 Joining Tables

 The mapunit table will at least provide a name corresponding with each mapunit.  To join information from the mapunit table to the attribute table of pecossoil (or whatever you named your clipped soil layer), right click the layer name for the context menu and select Join and Relate > Join.

 This many-to-one join is simple as shown by the dialogue boxes below.  Click the ‘About joining data’ button to learn more.

 After completing the join, note the new fields in your attribute table.
  There is still a lot more information that can be gleaned from this database if we relate the attribute table to the comp and layer tables.  Each mapunit is made up of a number of components.  If we relate the soils layer to the comp table we will have a  one-to-many relate where one record in the mapunit table is related to many records in the component table.

Spatial Averaging Soil Properties using Comppct

The average properties of soils in a particular map unit can be determined by using a weighted average of the component properties where the weights are equal to the percentages contained in the comppct field. For example, map unit NM572 contains soils with two surface textures: L (loam) for component 1 and GR-L (gravel loam) for component 2. The percentage of the total area in the map unit occupied by these components is 50% and 50%, respectively.

 To be turned in: Describe the NM963 soil in terms of its properties in the Mapunit and Component Tables. How many components does it have? What are their names? What percentage of the map unit does each component comprise? What is the predominant surface slope where this soil unit is found? What is the dominant soil texture? What percentage of the soil is in hydrologic soil groups A, B, C, D? Do these soil properties make sense considering where this soil is located?

 Next you should relate the component table (comp) with the layer table (layer).

 Layer Properties

The component properties describe the soil as a whole. Soils are usually divided into soil horizons beginning at the surface, and each horizon or layer has different physical properties. These are contained in the Layer table.

In the Table of Contents window, click on Source tab so that you can see your tables.  Right click the Layer table to access the properties.  In the Table Properties, with the Fields tab, note that you can choose which fields to show. Click off all the field names except the following: Muid, Seqnum, Layernum, Laydepl, Laydeph, Awcl and Awch (Awc is located near the end of the table). These properties represent:

·        Muid and Seqnum - the mapunit and component identification numbers defined in the component table.

·        Layernum - an integer between 1 and 6 that describes the number of layers, with layer number 1 being the uppermost layer.

·        Laydepl - the depth of the top of the layer in inches.

·        Laydeph - the depth of the bottom of the layer in inches.

·        Awcl - a lower limit on the estimated water holding capacity in inches of water per inch depth of soil (e.g. a value of 0.16 in/in means that 16 per cent of the soil volume is void space that could be occupied by water).

·        Awch - an upper limit on the estimate water holding capacity.

Relate the Layer and Comp tables together. Use the same procedure as before as you used to relate the mapunit and comp tables.

The mapunit table should now be related to the comp table and the comp table to the layer table. Click on Muid NM963 in the mapunit table and once it is highlighted, you should see the corresponding components and layers highlighted. If you don’t, you’ll need to select Options in the bottom right hand corner of the table, click Related Tables and the name you gave the relationship.

Vertically Integrating Soil Properties Over the Layers

Soil Depth The total depth of the soil layer is equal to the value of laydeph for its deepest component. For example, for map unit NM963, Component 1 has depth 0 - 9 inches for layer 1, 9 - 18 inches for layer 2, and 18-22 inches for layer 3 so the total depth is 22 inches for this component.

The average depth can be found as:

Average Depth = Sum over all components of (comppct/100) * component depth

Available Water Capacity

In the layer table, the Available Water Capacity is given in dimensionless units of inches of water / inch of soil. In soil water analysis, we need to know the Water Holding Capacity of the soil, that is, the total depth of water that the soil could store within its whole vertical profile. To calculate this, we first find an average value of available water capacity for each soil layer by averaging the low and high values for that layer, (awcl + awch)/2, multiply the result by the layer thickness (laydeph - laydepl), and sum over all the layers:

Water Holding Capacity of each Component = sum over its layers of (awcl + awch)/2 * (laydeph - laydepl)

This gives a result in inches of water. For example, for mapunit NM963, Component 1 is a Regnier clay loam having three layers, with awcl = 0.18, 0.14, and 0.00, respectively and awch = 0.20, 0.16, and 0.00. The average awc = (0.18 + 0.20)/2 = 0.19 for layer 1; (0.14+0.16)/2 = 0.15 for layer 2 and 0.00 for layer 3. Layer 1 is 9 inches thick so its water holding capacity = 0.19 * 9 = 1.71 inches, and similarly, Layer 2 being 18 - 9 = 9 inches thick, has a water holding capacity of 9 * 0.15 = 1.35 inches of water. Layer 3 has no water holding capacity.  The total water holding capacity of the Regnier Clay Loam component soil is 1.71 + 1.35 = 3.06 inches of water, whose storage is distributed throughout the pore spaces of the top 18 inches of soil.

The average value of the water holding capacity of the soils in the NM963 map unit is found as a weighted average of the values in each component using the comppct percentages as weights:

Average Water Holding Capacity for the Map Unit = Sum over all its components of (comppct/100) * Water Holding Capacity of each Component

Computing the average water holding capacity for a map unit is a fairly involved calculation so it is probably best to transfer the data to Excel and do the calculations there. If you click on the Options button at the bottom of a table, you can go to the Export Menu option and export the selected records of that Table into a new Table. If you choose the .dbf in the Export dialog you will produce a dBase file that can be read directly by Excel. Exporting and transferring the required records from the comp and layer tables to a spreadsheet may save you time in transcribing the data values manually.

To be turned in: For mapunit NM963, how many layers does each component have? What is the total soil depth (inches) for each layer and the average depth (inches) for the map unit? What is the total water holding capacity (inches of water) over the full soil depth for each component? What is the average water holding capacity (inches of water)for soils in this map unit?

 Land Use

Land use files are available from the USGS ftp site and from the EPA ftp site.  Additionally, EDAC makes land use files available.  There are a total of 23 coverages to makes up the state of New Mexico.  Two coverages are required for the headwaters of the Pecos:  Ftsumner and Santafe.  I have taken both of these coverages, projected them, and created export files.  Import these files into your ArcMap document.

From the RGIS website (rgis.unm.edu), select Data > Land Use/Land Cover.  Remember that you must click the folder (not the name) to see it expand.  We want to use 1:250,000 Quad USGS Land Use/Land Cover data.  It would be helpful know which quads cover the area of the pecos watershed.  Note that you can click the reference map at the top of the RGIS page.  You should find out that you need the Fort Sumner and Santa Fe 1:250,000 quads.  Download them (If you are not using an ArcInfo license of ArcGIS, it would be best to use a shapefile) and add the layers to your data frame in ArcMap.

These files use an Anderson Land Use Code classification system, in which major land use types are broken out into 9 categories:

1 = urban

2 = agriculture

3 = rangeland

4 = forest

5 = water

6 = wetlands

7 = barren land

8 = tundra

9 = ice and snow.

The second digit distinguishes subcategories of these principal categories, e.g.

11 = urban residential

12 = urban commercial

13 = urban industrial, etc.

This land use classification for the United States was made in the late 1970's and land use has changed in the years since then, particularly as cities have grown. But, the LULC files are still the standard land use classification of the United States taken as a whole.  Change the symbology of the layers so that they show the Anderson Land Use codes.  Below, I am changing the symbology for Fort Sumner.  I’m classifying by LUCODE.  I switched the Classification Method to Equal Interval with 8 classes and then manually entered the Break Values.

Once I entered the break values as shown below, the classification method changed to Manual.

 Next, you should change your labels and colors appropriately.

You should now feel pretty comfortable with the way that one of your land use layers appears.  Next, we want the other land use layer to look the same.  From the Symbology tab of the other land use layer, select import as shown below so that both land use layers will have the same legend.

Next, we’ll use some more features of the ArcToolbox.  You'll see a variety of geoprocessing tools by going to ArcToolbox>Data Management Tools>General.  Look at both 'merge' and 'append'.  Either one should work for our purposes.  I encourage you to read about both in Help.  If you use 'merge', you will leave existing feature classes unaltered, and create a new feature class.  'Append' works much the same way except for you will alter one data set rather than creating a new one.  Because you are altering a data set, the type of data set (i.e. coverage, shapefile, geodatabase) must align with your license level.  If you are working with an ArcView license level, then you need to use shapefiles or geodatabase feature classes.  If you are working with an ArcInfo license level, any of the formats can be used.

Once the land use has been merged or appended then 'intersect' the land use and soils.

Look at the attribute table for your new shapefile.  You have both soils information and land use information.  It would not be too difficult to go a step further and get curve number (CN) information (something some might do for their project).

 Once again, you should change the symbology of your final shapefile or feature class.

 
 Create a couple of layouts. 

The following is OPTIONAL!  It is not necessary to go through the command line ArcInfo work.

Following is an ARC/INFO session using STATSGO.  ARC/INFO allows you to add new items to tables based on values in other tables; whereas in ArcView, you may calcuate records in a new field based on other values in the same table.

TYPE NAME INTERNAL NAME NO. RECS LENGTH EXTERNL

------------------------------------------------------------------------------

DF NM.TIC ARC0000DAT 1343 12 XX

DF NM.BND ARC0001DAT 1 16 XX

DF NM.PAT ARC0002DAT 2084 24 XX

DF COMP ARC0003DAT 2754 292

DF COMPYLD ARC0004DAT 3252 60

DF FOREST ARC0005DAT 3247 24

DF INTERP ARC0006DAT 71523 24

7953 Records Selected.