USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Assessment of a Mallard Model in
Minnesota's Prairie Coteau

Appendix B

Determination of date-specific Robel readings for Conservation Reserve Program, wildlife planted cover, roadside right-of-way, unmanaged grass, and hayland habitat types, for 3 southwestern Minnesota study sites, 1990-92.

The mallard model (MM) uses, as variable inputs, 5 date-specific Robel readings (Robel et al. 1970) for each habitat type. These values specify a step function. Each step in the function represents cover height during a time increment within the mallard nesting period (March-August) (Mack 1991). Robel readings indicate cover height and density, which the MM uses as an index to relative attractiveness of habitats to nesting mallards on that date. The default readings are intended to portray long-term average habitat conditions and not necessarily those under which our study was conducted. We rescaled the default Robel readings in selected habitats in an attempt to more closely reflect cover height that existed in our study area during 1990-1992.

We obtained Robel readings measured in June from western Minnesota for Conservation Reserve Program habitat (CRP) and U.S. Fish and Wildlife Service Waterfowl Production Area habitat (WPA) in 1989-1991 (H. Kantrud, U.S. Geol. Survey, unpubl. data), and for roadside right-of-way habitat (ROW) in 1991 and CRP in 1992 (R. Welsh, Minn. Dep. Nat. Resour., unpubl. data) (Table B-1). We considered WPA habitat to be representative of the native and exotic grasses and forbs planted as cover on Wildlife Management Areas and Waterfowl Production Areas (i.e., wildlife planted cover). These values were compared with their respective MM defaults for June to determine their proportional differences from the defaults (%- DIFF). We then rescaled the step functions by multiplying each default value in a step function by the appropriate %-DIFFs (e.g., Fig. B-1) (Table B-2).

Table B-1. June Robel readings (dm) for Minnesota Conservation Reserve Program habitat (CRP), Waterfowl Prooduction Area grassland habitat (WPA)a, and right-of-way habitat (ROW), 1989-92b
  West-centralc South-centrald
1992     6.18  
1991 5.47 3.29   6.10
1990 3.54 1.72    
1989 4.08 2.10    
Avg. 4.36 2.37    
aWPA habitat assumed equivalent to wildlife planted cover in the Mallard Model.
bJune default Robel readings (dm) (Mack 1991) for comparison: CRP - 4.00, WPA - 4.00, ROW - 2.00.
cH. Kantrud, U.S. Geol. Survey, unpubl. data.
dR. Welsch, Minn. Dep. Nat. Resour., unpubl. data (includes Cottonwood and Jackson counties).
eCP1 mix of cool season grasses and alfalfa.

Table B-2. Comparison of dafault and customized Robel dates and readings (cm) for Minnesota Conservation Reserve Program habitat (CRP), wildlife planted cover (WMA), right-of-way (ROW), unmanaged grass, and hayland, 1990-92.
  Default values Customized values
Habitat type Date Reading Date 1990 1991 1992
CRP Apr 15 15 Apr 15 13 21 23
May 30 40 May 30 35 55 62
Jun 9 40 Jun 9 35 55 62
Jun 19 40 Jun 19 35 55 62
Jun 29 40 Jun 29 35 55 62
WMA Apr 15 12 Apr 15 6 12 13
May 30 40 May 30 17 33 34
Jun 9 40 Jun 9 17 33 34
Jun19 40 Jun 19 17 33 34
Jun 29 40 Jun 29 17 33 34
ROW Apr 15 9 Apr 15 18 28 31
Jun 1 20 Jun 1 40 61 69
Jun 30 20 Jul 11ª 40 61 69
Jul 1 5 Jul 12ª 5 5 5
Aug 10 5 Aug 10 5 5 5
Unmanaged grass May 1 8 May 1 7 11 12
May 25 13 May 25 12 18 20
Aug 8 13 Aug 8 12 18 20
Aug 9 13 Aug 9 12 18 20
Aug 10 13 Aug 10 12 18 20
Hayland Apr 20 8 Apr 20 7 11 12
Jun 19 35 May 31ª 31 48 54
Jun 20 5 Jun 1ª 5 5 5
Jul 20 20 Jul 8ª 18 27 31
Jul 21 5 Jul 9ª 5 5 5
ªDates differ from defaults (Mack 1991).

Unfortunately, no WPA value was available for 1992 nor did we have ROW values for 1990 or 1992. To estimate the WPA value, we divided the average 1989-1991 WPA value by the average 1989-1991 CRP value to obtain a mean ratio of WPA to CRP Robel readings in June (WPA/CRP). We then multiplied WPA/CRP by the 1992 June CRP value to estimate a 1992 June WPA reading. To estimate the ROW readings, we divided the 1991 ROW value by the 1991 CRP value to obtain a ratio of ROW to CRP Robel readings in June (ROW/CRP). We then multiplied ROW/CRP by the 1990 and 1992 June CRP value to estimate a 1990 and 1992 June ROW Robel reading. The estimated 1992 June WPA Robel reading and 1990 and 1992 ROW values were used to calculate %-DIFFs for ROW and WPA, which were then used to rescale the default ROW and wildlife planted cover step functions (Table B-2). Because the MM presumes that ROW habitat is mowed (i. e., the Robel reading is lowered to 5 cm) and we assumed that different yearly growth would not affect the height of mowed ROW habitat, we used the original MM default readings for ROW habitat after the mowing date.

We did not have year-specific Robel measurements for unmanaged grass or hayland habitat types. Much of the unmanaged grass on our study sites was ungrazed or lightly grazed and vegetation height was substantially taller than that depicted by the MM defaults for unmanaged grass. Thus, we determined Robel readings for unmanaged grass by averaging MM default readings for CRP and unmanaged grass. Further, we assumed that yearly vegetative growth in unmanaged grass and hayland would be affected by weather in a fashion similar to growth of CRP. Therefore, we used the yearly CRP %-DIFFs to rescale the step functions for unmanaged grass and hayland habitat types (Table B-2).

We also changed 6 dates in the step functions, which reflected mowing of hayland and ROW habitat, to ones suggested by wildlife managers (Table B-2).

Previous Section -- Appendix A
Return to Contents
Next Section -- Appendix C

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 19:34:33 EST
Reston, VA [vaww55]