STEP 3: SELECT CRITERIA FOR STRATIFYING THE STUDY AREA (IF NECESSARY)

In this step you will identify whether strata are necessary and, if so, which strata or different types of land or water body types will be used for your design and begin filling out the Sample Design Table. Specifically, you will identify which strata you will use, how many sample points will be collected in each, and the amount of resource that will be represented by each stratum. 

  • Stratification can be used to distribute sample points across the landscape or resource and/or to ensure that areas of interest, including reporting units, are sufficiently sampled (i.e., have adequate sample sizes for reporting). Stratification considers properties of the study area like physiography, management boundaries, ownership, or other attributes of the resource that need to be described to meet the monitoring objectives. Stratification decisions should be captured in Sample Design Table. 
  • The design process will typically start with the creation of a simple, minimally stratified, design across a broad area (e.g., LUP/RMP). That “draft” design will then be reviewed by the project lead and ID team to determine if the design is adequate or if different point allocations are necessary in certain areas. If more points are needed in specific areas, you may then add an intensification to the design in the future to ensure that you will obtain enough necessary information within those areas. 
  • Additional strata may be included in the design if deemed necessary. However, adding strata should be done with considerable thought, as sample sizes, required resources, and the complexity of data analysis increase with each additional stratum. Justify in 1-2 sentences why you have chosen your particular stratification scheme – do you strata add to your ability to answer your monitoring objectives? 
  • Additional stratification or point allocation approaches include but are not limited to:
    • Resource Management Plan boundaries 
    • Strahler stream order categories 
    • Habitat areas for sage grouse or other species of special concern such as T/E fish species 

 Terrestrial Designs 

  • The general recommendation for terrestrial monitoring designs is to not stratify. If stratification is necessary, the recommendation is to stratify by physiographic propertiesNote that physiographic properties are not typically used as reporting units. 
    • Stratifying by physiographic properties also helps allocate sample points to underrepresented or more variable portions of the landscape without sacrificing the ability to describe the whole landscape. 
  • Terrestrial monitoring has often stratified by LANDFIRE biophysical setting (BpS) groups, a remote sensing-derived layer that is conceptually very similar to NRCS Ecological Sites but is available as a continuous and consistent layer across the western US and therefore is commonly used in terrestrial sampling designs. BpS groups represent natural vegetation potential on the landscape based on biophysical environment and historic disturbance regimes. For more information, see the page on stratifying using LANDFIRE BpS. Other biophysical strata may be preferable in some cases. If you are planning to use alternate biophysical strata, you will need to create a GIS layer that spatially displays the stratification scheme and identifies the stratum names in the attribute table. This layer needs to be shared with the NOC and partners.
    • For example, if you are grouping Ecological Sites, please send a polygon shapefile of the Ecological Sites that you grouped together either already “dissolved” and named by group, or with an attribute field containing the stratum name that each Ecological Site belongs to. 
  • To identify the design strata, examine the GIS layer that you plan to use to develop your strata (e.g. LANDFIRE Biophysical Setting) in GIS and determine how many different types of terrestrial ecosystems exist within your study area.
    • If you have more than 10 different types in the study area, you may need to combine some of these ecotypes into groups in order to keep the design simple and manageable. For example, you may want to combine all BpS groups which are dominated by Wyoming Big Sagebrush into a single stratum. Document these groupings in the Stratification Lookup Table found in the appendix of the monitoring design worksheet.
      • Often, several different types of land that individually make up a small portion of the landscape will be grouped into an “Other” category to avoid inflating the number of points required by the design. If any of the strata are less than 3,000 acres or 1% of the study area, the NOC recommends that you group them with other strata so that the resulting stratum is greater than 3,000 acres or 1% of the study area. 
    • If you group several polygons to obtain your final strata, be sure to document how you made those decisions, and which polygons were combined to create the groups.

Lotic Designs 

  • The general recommendation for stream and river monitoring designs is to limit the use of strata unless we will not meet minimum sample sizes to report on specific areas or species of interest. However, all designs will be stratified by Strahler Stream Order, grouped into three categories:
    • Lower 48: small streams (1st and 2nd order), large streams (3rd and 4th order), and rivers (5th order and above).    
    • Alaska: small streams (1st order), large streams (2nd and 3rd order), and rivers (4th order and above).    
    • If any of the stream or river strata contain less than 1% of the total stream kilometers or result in less than three sample points, we recommend grouping that stratum with another stratum. 

Step 3 Example: Stratification

Step 3: Select criteria for stratifying the study area (if necessary)

The geospatial data layers used to define strata included:

  • Field Office boundaries
  • Terrestrial: LANDFIRE Biophysical Settings (BpS)
  • Streams/rivers: Strahler stream order categories from the NHD+, medium resolution version 2.0

Terrestrial

The terrestrial landscape of interest is variable, ranging from blackbrush plains to 10,000-foot mountain peaks. Terrestrial monitoring will utilize LANDFIRE Biophysical Settings (BpS) as strata to distribute sampling effort across the landscape but not for reporting purposes.

Lotic

The streams and rivers within the study area will be stratified by three Strahler Stream Order categories: small streams (1st and 2nd order), large streams (3rd and 4th order), and rivers (5th order and above).

Helpful Documents and Links

Monitoring Design Worksheet

Stratifying using LANDFIRE BpS

LANDFIRE Biophysical Setting

Comments are closed.