Potential Wetland Soil Landscapes (SSURGO) Data
Set
By: John M. Galbraith, Virginia Tech; Sharon Waltman, Steve Campbell, and Robert Dobos; USDA-NRCS, supported by Bill Wilen, USDA-USFS and Jeanne Christie
Association of State Wetland Managers, and the Wetland Mapping Consortium (WMC).
Purpose: The purpose of this document is to describe simple methods for using the Potential Wetland Soil Landscapes (PWSL) (Soil Survey Geographic Data Base) (SSURGO). SSURGO Metadata and User Guides are found on-line. The PWSL data set was intended for identifying sites that have a “hydric” soil as dominant component but are not mapped as National Wetland Inventory (NWI) wetlands. Areas with dominantly hydric soils but not recognized as wetlands may be because: a) the soil series meets hydric soil criteria 2 but not the hydric definition, b) the area is incorrectly mapped on the soil survey or the NWI maps, c) the area is missing wetland hydrology, or d) the area is not dominated by wetland vegetation. The PWSL data identifies soil landscapes that may be easily, inexpensively, or reliably restored to wetlands or converted to wetlands. The advantage of PWSL is that it is in raster format for spatial analysis and is a seamless set that does not require that users download SSURGO from each soil survey area and then join and register the data before use. Connectivity to other wetlands is not considered, although the data set may provide some information to be used in connectivity analysis (http://water.epa.gov/lawsregs/lawsguidance/cwa/wetlands/index.cfm; http://www.usace.army.mil/cecw/pages/reg_supp.aspx).
The PWSL data set includes the raster SSURGO spatial layer plus associated tables that contain data fields about hydric soils, hydrology, flooding, and ponding. PWSL data can be used for federal, state, and local programs and planning, wetland mapping, and potential wetland landscape inventory. Areas identified as dominantly hydric but not on NWI then require additional geospatial analysis and field investigation to confirm their potential as wetland landscapes. For example, a high resolution DEM produced from LiDAR may be used with a wetness index to find surface flow accumulation on the landscape. Additional geospatial data sets can be added in GIS software to produce a myriad of products.
Background: When wetlands are converted into a non-wetland land use under permit, the permitted person(s) must replace the wetlands or pay into a Restoration Fund. However, it may be difficult for the permitted person(s) to find a suitable site to mitigation the loss of wetland acres and functions. Normally the “mitigation” sites would be current wetlands that have been converted into a different land use by plowing, logging, or change in dominant vegetation, or former wetlands that have been modified by draining, excavating, filling, or flood protection. The areas of interest in this project are those areas where restoration of a current or former wetland to wetland functionality entails a simple plan with a high likelihood of success. For example, farmed wetland pastures need only removal of competitive plants and re-establishment of dominant hydrophytic vegetation to regain wetland status. Filled wetlands may also need expensive removal of fill material along with careful soil and vegetation reclamation.
The PWSL data consists of a 10-m and 30-m seamless gridded digital map layer of the 48 contiguous United States derived from SSURGO spatial and tabular data sets downloaded from Soil Data Mart (SDM) in Dec. 2009. Soil Data Mart also has individual state-wide datasets that can fit into an Access® Template downloaded from the SDM. SSURGO data (and associated Hydric and Wetland related attributes) is not readily available nationwide as a single shapefile. Also, the SSURGO vector format is not always desirable for merging with other data layers. Sharon Waltman, Norman Bliss (USGS) and others have developed a gridded version in two pixel sizes of SSURGO spatial data for the entire United States. These data can be shared with numerous federal agencies. The intent of USDA-NRCS is to refresh the spatial data every two years and update the queries whenever table, field names or data relationships change. The queries can be linked to The Wetlands Mapping Consortium Scholar Site hosted at Virginia Tech and the Association of State Wetland Managers (ASWM) Wetland Mapping website, and ideally cross-linked to the USDA-NRCS Soils web site.
Discussion: The simplest query of raster soil survey data would be to find all contiguous clusters of map unit pixels that have “hydric” dominant component but do not intersect with an NWI polygon. The clusters should have some additional data that indicates the site formed under wetland hydrology conditions.NHD or SSURGO point or line feature to indicate water presence or former presence, or protection from flooding or ponding. Combinations of layers may identify soils that have hydric soils and hydrology (indicated by drainage class) but have been cleared of hydrophytic vegetation (based on current land use versus climax vegetation type), have shallow fill, are artificially drained or protected from flooding. SSURGO map unit phases such as: high water table, poorly drained, slightly wet, and drained may be useful. Depressional and salt flats are water miscellaneous areas that may be helpful to identify.
If a pixel from a SSURGO soil map unit
has 50% or more hydric soils, but is not underlying an NWI polygon, it may be a
former wetland. This is especially likely if the land use is cropland or
pastureland. These areas may have high potential as restorable wetlands. The
area is less easily restored if the land use is now urban (assuming an accurate
Land Use layer). Comparison of current topography with former topography or an
on-site visit is needed. These areas may have moderate potential as restorable
wetlands. Other areas would have low potential as restorable wetlands.
Forest and pasture may be former wetlands if ditches or other conveyances are found to intersect hydric soil units with 50% or more hydric soils. The soil data includes tables that give supporting information about why the soil was rated as hydric, the hydroperiod, and soil properties that may indicate what type of wetland may be produced by the restoration. Some map unit names give information such as that can be useful. A map unit with 50% or more hydric soil that has “drained” or “ditched” or “protected” in the map unit name has potential to be easily restored. Ancillary data that also increases likelihood of the soil being a former wetland include hydrology data from NHD (intersection of part of the map units that have 50% or more hydric soil components with NHD marshes, swamps, springs or canal/ditch, connector, underground conduit and perennial stream segments indicate that hydrology may be easily restored). If NHD data is not available, intersection with SSURGO point data that has a marsh, wet spot, ditch, canal, spring, or other such special symbol may be used. The query contains the “mukey” (mapunit key) field, which can be used to join the Access table to the SSURGO spatial data. The simplest query would be to find all map unit pixels that have 50% or more hydric soil components, are not within an NWI polygon, and may have some NHD or SSURGO feature to indicate water presence or former presence, or protection from flooding or ponding. Combinations of layers may identify soils that have hydric soils and hydrology (indicated by drainage class) but have been cleared of hydrophytic vegetation or have shallow fill or are artificially drained or protected from flooding.
If a pixel from a SSURGO soil map unit
has 50% or more hydric soils, but is not underlying an NWI polygon, it may be a
former wetland. This is especially likely if the land use is cropland or
pastureland. These areas may have high potential as restorable wetlands. The
area is less easily restored if the land use is now urban (assuming an accurate
Land Use layer). Comparison of current topography with former topography or an
on-site visit is needed. These areas may have moderate potential as restorable
wetlands. Other areas would have low potential as restorable wetlands.
Forest and pasture may be former wetlands if ditches or other conveyances are found to intersect hydric soil units with 50% or more hydric soils. The soil data includes tables that give supporting information about why the soil was rated as hydric, the hydroperiod, and soil properties that may indicate what type of wetland may be produced by the restoration. Some map unit names give information such as that can be useful. A map unit with 50% or more hydric soil that has “drained” or “ditched” or “protected” in the map unit name has potential to be easily restored. Ancillary data that also increases likelihood of the soil being a former wetland include hydrology data from NHD (intersection of part of the map units that have 50% or more hydric soil components with NHD marshes, swamps, springs or canal/ditch, connector, underground conduit and perennial stream segments indicate that hydrology may be easily restored). If NHD data is not available, intersection with SSURGO point data that has features that indicate wetness or removal of wetness or flooding protection, such as marsh, wet spot, ditch, canal, spring, etc.
The data may be aggregated by LRR, MLRA,
state, county, or watershed boundary. The included SSURGO
tables provide a subset of SSURGO attributes needed by planners, developers,
conservationists, and managers. A description of the included tables is
provided in a later section of this document. They can be used for the:
Wetland Reserve Program (WRP),
Potential Wetland Soil Landscapes Project (PWSL)
National Wetland Inventory (NWI) (F&WS)
Overlaying the gridded SSURGO with the 2006 National Land Cover Data (NLCD2006) impervious surface (urban land) will provide an added benefit for watershed and wetlands managers.
Other important data sets include imagery data such as Landsat and NAIP,
National Agricultural Statistics Service (NASS) 2010 Cropland Data Layer (CDL),
National Hydrography Data set (NHD),
TIGER data,
Major Land Resource Area (MLRA),
Drainage Districts and HUC Watershed map layers
Specific examples of use: Please see Appendix A showing queries for Wetlands Reserve Program, and Appendix B showing queries for the Potential Wetland Soil Landscapes Project.
Contacts:
Bill Wilen Bill_Wilen@fws.gov National Wetlands Inventory
Wetlands Mapping Consortium - Jane Awl jane_awl@earthlink.net, Megan Lang (USFS) megan.lang@gmail.com, or John Galbraith john.galbraith@vt.edu
ASWM (Jeanne Christie or Sharon Weaver) http://aswm.org/wetland-science/wetland-mapping for links to the data and a user’s guide (future endeavor).
Potential Wetland Soil Landscapes Project - John Galbraith john.galbraith@vt.edu
Drainage Districts Maps, Improved land Use Land Cover Data – Anne Neale Neale.Anne@epamail.epa.gov
Wetland Reserve Program Query - Steve Campbell Steve.Campbell@por.usda.gov
Potential
Wetland Soil Landscapes (SSURGO) Data Set - Sharon Waltman
Sharon.Waltman@wv.usda.gov
___________________________________________________________________________
APPENDIX A: Wetlands
Reserve Program
An MS Access®
query was created to extract data from the Soil Data Mart core data set that
would be useful in identifying potential sites for the Wetlands Reserve Program
(WRP). It aggregates data by mapunit for
spatial display. The query contains the
“mukey” field, which can be used to join the Access table to the SSURGO spatial
data. Pipe delimited files imported into
an Arc File Geodatabase (FGDB) offers a good solution
for tables that have greater than 65,000 records (national component and map
unit tables).
The following
are tables and fields used to create the MS Access® database
“US_WRP_potential_SDM_2009.mdb” in the table “US_WRP_Potential”.
Spatial Table – Mapunit Key (a.k.a. mukey)
Legend table- areasymbol - soil survey area symbol
Legend table- areaname – soil
survey area name
Mapunit table- musym - mapunit component key (a.k.a. cokey)
Mapunit table- mukey – mapunit key
Mapunit table- muname - mapunit name
Component table- hydricpct* - percent of the mapunit that
consists of hydric components
Comonth table- freqfloodpct* - percent of
the mapunit that consists of components that have frequent or very frequent
flooding frequency during any month.
Cosoilmoist table- wattab24pct* - percent of the mapunit that
consists of components that have a water table (wet state) within 24 inches (61
cm) of the surface during any month
Chorizon table- ksatslowpct*
- percent of the mapunit that consists of components that have slow or very
slow Ksat (<= 1.41 um/sec RV) within 24 inches (61 cm) of the surface. A subset
of this table was created that contained only the Ksat
data and field needed to join it to the component table (cokey).
Component table- hsgDpct* - percent of the mapunit that
consists of components assigned to hydrologic soil group D (or A/D).
Developed by:
Steve Campbell
Soil Scientist
USDA - Natural Resources Conservation Service
West National Technology Support Center
1201 NE Lloyd Blvd., Suite 1000
Portland, OR 97232-1208
Phone:
503-273-2421
E-mail: steve.campbell@por.usda.gov
___________________________________________________________________________
Appendix B: Potential Wetland Soil Landscapes (SSURGO) Data Set
A Geographic
Information System (GIS) software program should be used to identify clusters
of two or more contiguous pixels (share one or more sides with similar
identified pixels) with “hydric” dominant soils from the spatial data set. The
clusters should not intersect with NWI pixels.
An MS Access®
query will be created to extract tabular data from the Soil Data Mart core data
set by state that
would be useful in identifying potential sites for the Wetlands Reserve Program
(WRP). It aggregates data by mapunit for spatial display. The query contains the “mukey”
field, which can be used to join the Access table to the SSURGO spatial data. Pipe delimited files imported into an Arc
File Geodatabase (FGDB) offers a good solution for
tables that have greater than 65,000 records (national component and map unit
tables).
Because of the enormous size of the database for the entire
48 contiguous United States, only a subset of the data is requested. The following
are SSURGO
2.2.5 tables and columns
(fields) needed to create the MS Access® database “US_PWSL_Potential_SDM_2009.mdb” in the table “US_WSL_Potential”.
The codes
are available at: link . [JG1] Additional
SSURGO data can be downloaded for individual soil survey areas through the Soil
Data Mart.
Spatial Table (area) – (mukey)
Spatial Table (point) - Point data that indicates dams, levees, ditches, canals, marsh or swamp, miscellaneous water, perennial water, wet spots, saline spots, closed depression, streams, springs, or flood pool line. Note:These may be found in Feature Description table (areasymbol, featkey, featsym, featname, featdesc), Feature Point table (areasymbol, featkey, featsym) and Feature Line table (areasymbol, featkey, featsym)
Cohydriccriteria table – (cokey, cohydcritkey, hydriccriterion)
Comonth table – (cokey, comonthkey, flodfreqcl, floddurcl, floddurcl, ponddurc, ponddep_r)
Cosoilmoist table – (comonthkey, comonthkey, soimoistdept_r, soimoistdepb_r, soimoiststat)
Component table – (mukey, cokey, majcompflag, compcomppct_r, hydricrating, drainagecl, hydgrp)
Legend table – (areasymbol, lkey)
Mapunit table – (mukey, lkey, musym, muname)
Find the soil map unit pixels that have the dominant soil with hydricrating = y.
To do this, use the following query from the full data set.
select areasymbol, musym, compname, mapunit.mukey, comppct_r, from legend
join mapunit on legend.lkey=mapunit.lkey
join component on component.mukey = mapunit.mukey
join cohydriccriteria on component.mukey = mapunit.mukey
where areasymbol like 'pa%' and
areasymbol <> 'US' and
hydricrating = 'y' and
majcompflag = 'yes' and
component.cokey in
(select top 1 c.cokey from component c where c.mukey=mapunit.mukey
order by comppct_r desc)
Developed
by:
Sharon W. Waltman, USDA-NRCS
3040 University Avenue, Suite 3037
Morgantown, WV 26505
Phone: 304-293-9835
E-mail: sharon.waltman@wv.usda.gov
Robert Dobos, USDA-NRCS
100 Centennial Mall North, Room 152
Lincoln, NE 68508-3866
Phone: 402-437-4149
402-617-9853 (cell)
E-mail: bob.dobos@lin.usda.gov
The initial Research and Development data will then be made available upon request through the Assoc. of State Wetland Managers[sww3] until posted on a national public domain server site.
Jeanne Christie or Sharon Weaver
Association of State Wetland Managers
(207) 892-3399 phone,
(207) 892-3089 fax,
(207) 310-8708 cell
website: http://aswm.org/
blog: http://aswm.org/wordpress/
Example
Methodology for Compiling the Geospatial Data Layers (10-m Raster SSURGO data,
statewide coverage) for Potential Wetland Soil Landscapes
The
following steps are an example methodology to create an Easily Restored
Wetlands Geospatial Information System and Database, using the Potential
Wetland Soil Landscapes (Easily Restored Wetlands) (SSURGO) Data Set, following
the steps and guides provided at: http://soildatamart.nrcs.usda.gov/SSURGOMetadata.aspx.
Step 1) Add the Potential Wetland Soil Landscapes (SSURGO) Data Set for your state or area of interest.
{Insert screen capture/s here}
Step 2) Find all map units or pixels where the dominant soil component is hydric.
{Insert screen capture/s here}
Step 3) Mask out all pixels that do not intersect with an NWI polygons. (call that the “hydric_not_NWI” layer).
{Insert screen capture/s here}
Step 4) Remove all pixels that are single, or only touch other pixels on an edge, leaving clusters of “hydric_not_NWI” pixels.
{Insert screen capture/s here}
Step 5) Identify clusters of “hydric_not_NWI” pixels that are not completely overlapping NLCD 2006 pixels of 23 Developed, Medium Intensity or 24 Developed High Intensity. Exclude those clusters.
{Insert screen capture/s here}
Step 6) Intersect clusters of “hydric_not_NWI_not_developed” pixels with the NLCD 2006 or an improved land cover map to identify the major land cover. Expert knowledge will guide selection of land cover(s) that might cause the hydric soil areas to not be identified on NWI maps because of vegetative manipulation. Call this layer “hydric_not_NWI_bc_vegetation_not_developed”.
{Insert screen capture/s here, maybe a table}
Step 7) Use expert knowledge to clusters with evidence that there has been hydrologic modification. For example, intersect the “hydric_not_NWI_not_developed” layer with NHD or SSURGO feature to indicate current water presence (marsh symbol) or former presence (ditches), or protection from flooding or ponding (dams and levees). This might include NHD marshes, swamps, springs or canal/ditch, connector, underground conduit and perennial stream segments. Intersection with SSURGO point data that has a marsh, wet spot, ditch, canal, spring, or other such special symbol may be useful. Also, a map unit that has “drained” or “ditched” or “protected” in the map unit name has potential to be a former wetland. Call this layer “hydric_not_NWI_bc_water_not_developed”.
{Insert screen capture/s here}
Step 8) Place a hierarchy (high, moderate or low potential as restorable wetland soil landscapes) on pixel clusters that are most likely to be wetlands or easily restored as wetlands in both the “hydric_not_NWI_bc_vegetation_not_developed” and “hydric_not_NWI_bc_water_not_developed” layers. For example, some areas may be very easily restorable by simply filling ditches, while others may be costly and have low potential for restoration success because of severe previous modification. Each state may choose how to prioritize their layers.
{Insert screen capture/s here, maybe an example hierarchy table}
Step 9) The data may be aggregated or subset by LRR, MLRA, state, county, or watershed boundary. For example, to build the the “NLCD_wetlands_hydric_not_NWI” layer that intersects an NHD “ditch” in Virginia, you would add layer “ “ in the ‘va” folder…..and then ……
{Insert screen capture/s and further steps here }