Explanation of Variables
For detailed explanations download the 2002 Methods Manual (.pdf)
POINTID: This is a unique identifier number for each exact point location. The numbers are arbitrary, but they are included here to indicate any movement of points between years. Points in later years are exactly the same only if the PointID number is exactly the same. Otherwise there was some adjustment in the location of the point on the ground, even if the transect and stop numbers are the same. The main reasons for relocating points are the adjustment of points that were too close together and the inability to find the point marker from previous years.
TRANSECT: A unique identifier for each transect based on the latitude and longitude of the start point, e.g. Transect # 453411412 would begin about 45 deg. 34 min. N; 114 deg. 12 min. W (not exact -- see GISLAT and GISLON for exact location for use in GIS).
STOP: The point position along the 10-point transect.
DATE: includes only the month and day (e.g., 603 = June 3rd). The YEAR has been moved into a separate variable with four digits.
COVTYPE (see 5-digit codes in Methods Manual): The 5-digit cover types collected in the field. The coding scheme became more detailed in later years, so codes from 1994 - 1996 may not match with current list. Also, field crews classified points anew each year and these often do not agree between years (we will eventually make one "consensus" file for all points).
COVER: (1994 - 1998) The numerous, 5-digit cover types (COVTYPE) have been consolidated into 18 general COVER types (a few are blank, where rare or ambiguous cover types were unassigned). This is the scheme used in the GTR (RMRS-32) publication:
'Cedar/Hemlock or Grand Fir'
'Spruce/Fir'
'Lodgepole Pine'
'Mixed Conifer'
'Douglas-fir'
'Ponderosa Pine'
' Young Forest'
'Partial Cut'
'Patch Cut'
'Seed-tree Cut'
'Clearcut'
'Old Clearcut'
'Post-Fire'
'Sagebrush'
'Grassland'
'Wetland'
'Riparian Shrub'
' Aspen' or ' Cottonwood'.
EDGE: O = only one cover type (from our list) within 100-m-radius circle; 1 = more than one cover type within 100 m. We typically exclude the latter for most habitat analyses because we do not know which cover type the birds at these 'edge' points were using. Edge designation was not always straightforward for field observers. For many heterogeneous areas, it is a judgment call as to whether sections within the 100-m-radius circle are different enough to be classified as different cover types (according to our scheme) or if it all represents finer-resolution heterogeneity within a single cover type. Of course, errors in estimating the distance to an edge (whether or not it is in the circle), or simply not seeing minor cover types, also are expected. If you compare between years, you will note that there was disagreement for many points on whether a point was within 100 m of an edge.
EDGE1 and EDGE2: 1995-2000 only. 5-digit codes for the 2nd and 3rd most important cover types within 100 m. If <3 observed, spaces filled with 99999. We started this procedure in 1995. Then, in 1996 we started to report, for each bird observation, whether or not the bird was in the "main" cover type designated for the point. These data are available but are not taken into consideration in the data files here. All birds within 100 m are included here. The cover type location of the birds cannot be used in quantitative analyses because we do not have the proportion of the circle that each cover type represents.
INDIVIDUAL BIRD SPECIES VARIABLES (see 4-letter codes in methods manual):
Total number of each species detected within 100 m at each point; flyovers are excluded for all species except: swallows, Vaux's Swift, Sprague's Pipit, Common Snipe, and Common Nighthawk. There may be a few rare species that were not included. For a complete list, request the raw bird data.
GISLAT/GISLON: 1994 only. This is the exact latitude and longitude where the point was placed on the GIS (in decimal degrees). It can therefore be used to place the points in ArcView or any other GIS. The points were placed on a GIS grid (by Wendy Williams at the Wildlife Spatial Analysis Lab, University of Montana) by comparing placement on aerial photos marked in the field. This has only been done for 1994. Points in later years will be exactly the same only if the PointID number is exactly the same. Otherwise, if the transect numbers are the same, transects are in the same location but exact point placement may be different. Be aware that we have not rechecked the placements of the points to determine why many of the cover types from the GIS do not agree with those recorded by field observers.
HEIGHT: is a bit of a mess! In 1994, the highest vegetation layer was estimated, down to zero, and we have returned to this method as of 1999. However, in 1995-1998, we recorded "999" if there was no canopy layer, so these must be removed before any summary statistics are calculated. Be aware that some observers apparently estimated "canopy height" down to 1 m and only called it "999" at zero, whereas others followed instructions by recording "999" for any stand that had not yet reached the closed-canopy stage.
Copyright© Spectral Fusion 2004. All Rights Reserved.