Field Protocols
MEASURING VEGETATION
We have established methods for sampling vegetation in forest, shrub, and grassland habitats. Exact methods differ somewhat between habitats. The current protocol describes only measurements in forest and shrub communities; at the time of writing (winter, 1997) grassland vegetation measurements are described in a supplemental protocol which will be incorporated into the main protocol in the future. Differences in sampling protocol among habitats will be noted, below. We use a modified version of the James and Shugart (1970) method for sampling vegetation in forest and shrub systems. Basic to this protocol is setting up circular plots in which to measure vegetation.
In this section, we first describe some of the techniques used to measure vegetation. Then we outline the data collected at all vegetation plots (both nest sites and sites used to systematically describe vegetation on a plot), organizing the data into 1) measurements taken at the nest, 2) measurements taken within 5 m plots (see below), and 3) measurements taken within 11.3 m plots (see below).
Locations Where Vegetation Should Be Measured
There are 3 types of locations where vegetation plots have to be established, and vegetation measured:
- OPEN and CAVITY NESTS -- 5- and 11.3-m radius plots centered on nests for all nests known to have contained eggs.
- SYSTEMATIC DESCRIPTION OF VEGETATION ON PLOTS -- A series of points in a grid system should be established to sample vegetation at the plot level. For sites that are doing bird point counts, the plot vegetation points should be centered on the survey points. Four pairs of 5- and 11.3 m vegetation plots should be done at each plot vegetation point. See next section for more details.
- VEGETATION ON NESTS WITHOUT EGGS -- Use your own judgment on whether or not to measure vegetation at nests that failed before laying or with unknown fate (status unknown/nest not occupied). If your nest numbers are low for the species in question then go ahead and take vegetation measurements. Don't bother with vegetation on these "status unknown" nests if sample sizes are large enough (e.g. if you already have 30 or more vegetation descriptions for a given species within the year).
Previous versions of the BBIRD protocol specified that vegetation plots also be established at paired locations for each nest for which vegetation was measured. These paired, "non-use" vegetation plots were intended to be used in order to determine whether birds were selectively placing nests in certain habitats, after any effect due to tree (or other substrate) type of nest had been eliminated. At the fall 1996 BBIRD meeting it was decided that continued measurement of vegetation at these paired, non-use, sites was not consistently providing us with the desired information. Hence, measurement of non-use plots has been discontinued as a required part of the BBIRD protocol. However, if researchers at an individual site can see the use of such data for their individual projects, we encourage them to continue measuring vegetation at these locations.
Spatial Arrangement of Systematic Sampling Plots
A stratified random sample of vegetation on nest search plots should be conducted every 3 years, except in early successional habitat in which vegetation should be measured more frequently. Instead of conducting a single large survey every third year, you should conduct vegetation measurements on 1/3 of the locations each year in a rotating fashion. For sites in which you will be conducting point counts, plot vegetation points will be established at each survey point. This gives both a measure of plot vegetation and allows us to tie point count data to the vegetative characteristics of the 50 m radius survey plot. Studies that are not doing point counts should establish a series of plot vegetation sampling points spaced 200 m apart, and 100 m from the edge of the plot (as described for survey point establishment). At each location, four separate vegetation plots should be done (see figure, below). The first plot will be located at the sampling location itself, and the remaining three will be located at 120 degrees from each other around the central location (the direction of the first outer subpoint should be random). The center of outer plots should be located 30 m from the center plot. Location of each of the four subpoints should be permanently marked with stakes to allow measurement of the same vegetation sampling plots in future years. Both 5 and 11.3 m vegetation plots should be done at each of the 4 subpoints as described for nest.
FIGURE: SPATIAL ARRANGEMENT OF SYSTEMATIC VEGETATION PLOTS
Setting Up An Individual Vegetation Plot
At most study sites, two sizes of circular plots are established: 1) a plot of 5 meter radius that is used to count shrub and sapling stems and measure ground cover, and 2) an 11.3 m radius plot that is used to count the stems of trees. The 5 m plot is nested within and centered on the same point as the 11.3 m plot.
Plots should be divided into quadrants to facilitate stem counting (see figure, below). Use a central tent stake attached to four 5 m long ropes to mark the center of the plot and divide the area into 4 quadrants; the ends of the ropes can be attached to the ground with stakes or tied off on vegetation. Using compass bearings is a convenient way to establish the quadrates in areas with little or no topography. However, on slopes you may find it simpler to orient one axis downslope along the plot's aspect, and the other axis parallel to the slope. For cavity nesters, having two arms of rope extent at 45 angles on either side of the cavity's orientation is convenient. Count stems in each quadrant separately. Setting up ropes and stakes takes little time but improves accuracy and efficiency.
FIGURE: 5 METER VEGETATION PLOT
Exceptions To Rules For Plot Size: Smaller Plots in Dense Habitat
Counting stems in the 5 m radius circle is extremely time consuming in some shrubby habitats. In thick, open canopy, shrub habitats dominated by very large numbers of small stems it is permissible to reduce the size of the 5 m circle to 1 m for stems < 2.5 cm in size. Continue to count 2.5 - 8 cm DBH stems in the 5 m circle. Do not use 1 m plots in closed canopy forest habitats. Be cautious in deciding to go to 1 m plots. Small plots may not adequately sample heterogeneous habitats. Stick with larger plots if 1 m plots do not appear to be representative of the 5 m plot. As a rule of thumb you should probably stick with 5 m plots unless you are getting more than 100 stems per 5 m quadrant. Be sure to record whether stems were counted in a 1- or 5-m plot.
In areas where blackberry or trailing vines form very dense impenetrable clumps with numerous vertical and horizontal stems, estimate percent of the ground covered by such patches instead of counting stems. This should be done in the 5 m radius circle, not a 1 m circle. For blackberry species that form single stem plants that can be counted discretely, continue to count stems. Use the guidelines in the previous section to decide whether it is permissible to use 1 m plots to sample single stem blackberry species.
Reading the densiometer
Densiometers are used to measures forest cover. The densiometer is a convex mirror with a grid etched into it (obtained from Robert E. Lemmon, Forest Densiometers, 5733 SE Cornell Drive, Bartlesville, OK 74006; (918)-333-2830 or Forestry Suppliers. We use the convex model). There are 24 squares in the mirror. Imagine 4 uniformly spaced dots in each square of the grid, as illustrated (see figure, below), and count either the number of dots covered by vegetation OR the number not covered by vegetation. When percent canopy cover is high, it is easier to count dots not covered and subtract the total from 96. When canopy cover is low, it is quicker to count covered dots and record that number. Always record number of dots covered, not number uncovered. Do not break the squares into 4 imaginary squares and try to decide if each one has greater or less than 50% cover. The 4 squares within a square method introduces bias because individuals tend to decide close calls more frequently in one direction than another. Deciding whether or not an imaginary dot is covered is a simple objective decision, and is less susceptible to bias. Avoid counting trunks as cover when possible, particularly when they make up a substantial portion of the cover. Step away from the trunk if necessary. Dot numbers should be converted to percent canopy cover (#dots covered/96*100) for reporting purposes, but field assistants should not do these conversions in the field.
FIGURE: DIAGRAM OF DENSIOMETER SURFACE
Measurements Made At From The Center Of Vegetation Plots
The following measures are taken while standing at the center of the plot (nest, point count marker, or grid marker). Note that these measures are divided into two groups. The first group of measurements are taken at all types of plots (nest, and systematic vegetation). However, the second group of measurements is only taken at plots used to systematically describe the vegetation on study plots. First, the following variables are measured at all vegetation plots:
- NEST TYPE (NSTTYPE)
- a variable that describes whether the subsequent nest and vegetation record belongs to a nest (code NSTTYPE as 1), a non-use plot (code as 2), or a systematic/point count vegetation plot (code as 3).
- AVERAGE TOP CANOPY HEIGHT (TOPCAN_M)
- Choose a point in the canopy that represents the average height of the top of the canopy within 11.3 m of the center of the plot (i.e. ignore lone trees that emerge above the main canopy when taking this measure). Measure the height of this point with a clinometer.
- TOTAL CANOPY (CANCOV)
- Percent total canopy closure measured with a densiometer. Stand at/under the nest and take 4 densiometer readings, turning 90° between readings. Measure both CANCOV and HCANOPY (below) at the same time, for each of the four directions. CANCOV is intended to measure total canopy cover, including sub-canopy cover.
- HIGH CANOPY (HCANOPY)
- Percent total canopy cover above 5 m in height. This measure is intended to measure the upper canopy, not lower canopy layers. Stand at/under the nest and take 4 densiometer readings, turning 90° between readings. This measure can be difficult if there is a lot of low vegetation. We want an estimate of high canopy cover so try the following methods to avoid low vegetation:
- Step away from the nest to a more open area that is representative of the high canopy above the nest.
- If there is relatively little low vegetation you can sometimes ignore it when counting densiometer dots covered and uncovered.
- If low cover is too continuous for a good densiometer reading then make an ocular estimate of cover above 5 m.
- DOMINANT PLANT SPECIES IN CANOPY (CDOMSP1)
- alphanumeric code (from PLANTS database; USDA 1994) of plant species that dominates the high canopy. Species' dominance is determined by eye. Record species that account for at least 40% of the high canopy present. Leave blank if no single plant species represents > 40% of the high canopy present.
- PERCENT OF DOMINANT CANOPY SPECIES (CDOMSP1P)
- percent of high canopy present that is occupied by CDOMSP1.
- CO-DOMINANT PLANT SPECIES IN CANOPY (CDOMSP2)
- alphanumeric code (from PLANTS database; USDA 1994) of plant species that co-dominates the high canopy. Use this variable when there are 2 plant species that each represent > 40% of the high canopy present. Leave blank if there is not a second plant species that represents > 40% of the high canopy that is present.
- PERCENT OF CO-DOMINANT CANOPY SPECIES (CDOMSP2P)
- percent of high canopy occupied by CDOMSP2.
This next set of measurements are taken at all plots, except vegetation plots at nests. These measures are all based on the point-centered quarter method of estimating densities of plants (e.g., Mueller-Dombois and Ellenberg 1974). For these measurements, stand at the center of the plot, and locate the nearest live tree, live shrub, or snag (dead tree) within each of the quarters of the circle surrounding you. It is convenient to divide the circle into quadrates along the cardinal compass directions. Within each quadrate, record the following information:
- SPECIES, CLOSEST SHRUB (PQSHSP#)
- where # is a number from 1 -- 4 that represents identifies the quadrate
- DISTANCE TO SHRUB (PQSHDST#)
- the distance (in meters) from the center of the plot to the shrub
- HEIGHT OF SHRUB (PQShHt#) height (in meters) of selected shrub
- MAXIMUM SHRUB WIDTH (PQShWd#) maximum width (in meters) of selected shrub
- WIDTH PERPENDICULAR TO MAXIMUM (PQSHPWD#) width of the shrub measured at a right angle to the maximum width
- SPECIES OF NEAREST TREE (PQTRSP#)
- species code for closest live tree within the quadrate, where quadrates are indicated by the # (1 -- 4) in the variable name
- DISTANCE TO CLOSEST TREE (PQTRDST#)
- distance (in meters) from the center of the plot to the closest tree within a quadrate
- HEIGHT OF CLOSEST TREE (PQTrHt#) height (in meters) of closest tree in quadrate # to the center of the plot
- DIAMETER AT BREAST HEIGHT, CLOSEST TREE (PQTrDBH#)
- diameter at breast height (in centimeters) of closest tree in quadrate # to the center of the plot
- AVERAGE CROWN WIDTH, CLOSEST TREE (PQTrCrn#) average width (in meters) of crown of closest tree in quadrate # to the center of the plot
- CANOPY COVER, CLOSEST TREE (PQTrCan#) canopy cover (from densiometer) under closest tree in quadrate # to the center of the plot
- DISTANCE, CLOSEST SNAG (PQSNAGD#)
- distance (in meters) from center of quadrate to the closest snag with 12cm dbh in quadrate #. Leave this and all subsequent measure of snag # blank if there is no snag present within 25 meters
- HEIGHT OF CLOSEST SNAG (PQSngHt#) height (in meters) of closest snag to center of plot in quadrate #
- AVERAGE CROWN WIDTH, CLOSEST SNAG (PQSngCr#) average width (in meters) of crown of snag closest to center of plot in quadrate #
- DIAMETER BREAST HEIGHT, CLOSEST SNAG (PQSnDBH#)
- diameter at breast height (in centimeters) of the clostest snag in quadrate # to the center of the plot.
- CANOPY COVER, CLOSEST SNAG (PQSnCan#)
- canopy cover (from densiometer) under the closest snag in quadrate #
Measurements Taken Within The Small (5 Meter) Plot
The following are the measurements to be taken within the 5 meter radius plots. All measurements are taken for all vegetation plots. One measurement taken is a count of the numbers of stems of shrubs that exist within the plot circle. Stems of all saplings and shrubs should be counted by species within each 5 m plot at 10 cm above the ground. The number of stems of each species should be counted for each of two size classes (<2.5 cm diameter or >2.5 cm diameter). We make no distinction in the data between shrubs and saplings, but different criteria must be used to place shrubs (often having no main stem) and saplings (often having a single, main stem) in one of the two size classes into which we place shrubs. Separate counts are made of the number of stems of each species of shrub/sapling within the plot. Please note: growth form and size class do not constitute 4 different categories. We are only categorizing stems as small or large, not as single stem small, multiple stem large, etc.. Counts the numbers of stems that fit any of these criteria:
- No single central stem at which DBH can be measured:
- Small Size Class: < 2.5 cm stem diameter at 10 cm above ground
- Large Size Class: > 2.5 cm stem diameter at 10 cm above ground
- With a single central stem
- Small Size Class: < 2.5 cm dbh, or less than 1.4 m tall
- Large Size Class: 2.5 - 8.0 cm dbh
Many plant species break into multiple stems fairly close to the ground. In these situations, it is reasonable to assume that birds respond to stem densities rather than individual plant numbers. Therefore, we count vertical stems, not individual plants.
FIGURE: RULES FOR STEM COUNTS
Rules for counting stems:
- Don't count plants/stems less than 50 cm (i.e. approximately knee height) high.
- Count the number of vertical stems at 10 cm above the ground (ankle level), i.e if a stem branches above 10 cm then it is counted as 1 (see figure, below)
In each of the 5m plots we also measure the depth of organic litter, using a stake or other tool (meter stick, ruler) to dig a small hole down to where individual leaf parts are no longer visible (leaf veins usually decompose last). We are interested in the depth of leaf litter and partially decomposed organic matter that accumulates on top of the mineral soil. Litter depth is measured at 2 m intervals along the ropes that mark the 5 m plots; marking the locations for litter measurements on the ropes will aid consistency. If any of these 12 points land on a log or a rock, move the meter stick slightly to a location where you are actually measuring litter depth.
FIGURE: LOCATIONS FOR LITTER DEPTH MEASUREMENTS
FIGURE: GROUND COVER MEASUREMENTS
The final class of vegetation measurements made within the 5 meter plots are estimates of ground cover, of several types. For each of the 4 quadrants in the 5 m plot, make an ocular estimate of: the percent of the ground covered by green vegetation from 50 cm above ground, to ground level. For each class of vegetation (shrub, grass, fern etc..), estimate a percent value independent of all other vegetation types (i.e. as if other vegetation types were absent). Vegetative cover categories can sum to more than 100% because of vertical stratification of plant layers. However, no single category of vegetative cover should be greater than the value for ALL GREEN COVER, GCGRN.
Ground cover measurements in the 5 m vegetation plot. Notice that there are two types of ground cover that affect cover estimation rules: tall sparse cover that can overlap with low cover types, and low dense cover that cannot overlap with other low cover types (see figure to left). Percent cover in high and low cover types can sum to more than 100%, but low cover types (low grass, bare ground, moss, low dense shrub, leaf litter, logs, etc..) alone must sum to 100%. Total green cover can be more than 100%.
The following are the variables to be measured within each 5 meter plot:
- PLOT ASPECT (ASPECT_D)
- The direction the 5 m plot faces in degrees. From the top of the 5 m plot face downhill and take a compass reading in degrees. For cavity nests also measure aspect at 5 m uphill from nest (not at top of 11.3 m circle).
- SLOPE (SLOPE_D)
- Measure the slope across the 5 m plot from the bottom to the top of the plot in degrees (LEFT HAND SCALE OF CLINOMETER). For cavity nests also measure slope across a 5 m radius circle centered on the nest tree (not across the 11.3 m circle). Measuring at eye-level aiming at a target also at eye-height the simplest way of taking this measurement.
- LITTER DEPTH (LITTR_MM)
- Organic litter depth (in mm) should be measured across the center of the plot parallel and perpendicular to the slope of the plot. Measures should be made at 12 points as shown in the figure, above. LITTR_MM is the average of these 12 measurements. We are measuring litter depth, not percent of ground covered by litter. Note if you are in an area covered in water, leave LITTR_MM blank, and do not record litter depth as zero: there is litter...just not at the surface.
- RADIUS, SMALL PLOT (VGSRADM)
- radius (in meters) of the small vegetation plot. This will generally be 5 meters.
- SHRUB/SAPLING SPECIES (VGSP#)
- species code for the #th (e.g., VGSP1, VGSP2, etc.) species of shrub encountered in the 5 meter plot. Species do not have to be placed in any specific order
- NUMBER OF SMALL SHRUB STEMS (VSTM#CL1)
- the number of small (<2.5 cm diameter) diameter stems of species # found within the plot circle
- NUMBER OF SMALL SHRUB STEMS (VSTM#CL2)
- the number of large (2.5 -- 8 cm diameter) diameter stems of species # found within the plot circle
Use as many VGSP#, VSTM#CL1, and VSTM#CL2 as required for the species of shrubs encountered. Rare species can be pooled into the group "OTHER".
- % ALL GREEN COVER (GCGRN) percentage of ground covered by green vegetation that is below 50 cm in height
- % GRASS COVER (GCNGRASS) percentage of ground covered by grasses below 50 cm in height. This variable is new, as in previous years cover from sedges and grasses were lumped into the variable GCGRASS.
- % SEDGE COVER (GCSEDGE)
- percentage of ground covered by sedge that is below 50cm in height.
- % SHRUB COVER (GCSHRUB)
- percentage of gound covered by woody perrenial plants that are below 50 cm tall.
- % BRUSH COVER (GCBRUSH)
- percentage of ground covered by small dead woody vegetation (i.e. dead shrubs and bramble) less than 50 cm above the ground
- % FORB COVER (GCFORB)
- percentage of ground covered by broad-leafed non-woody plants below 50 cm height.
- % FERN COVER (GCFERN)
- below 50 cm.
- % MOSS COVER (GCMOSS)
- percentage of ground covered by moss
- % CACTUS COVER (GCCACTUS)
- percentage of ground covered by cactus that are less then 50cm tall
- % LEAF LITTER COVER (GCLEAF)
- percent of ground covered by leaf litter. This value should be independent of taller, sparser vegetation (litter + tall sparse vegetation can sum to more than 100%), but is dependent on low dense vegetation (litter + low dense vegetation sum to 100% or less). Example: a plot with a layer of small shrubs/saplings covering 80% of the ground at 50 cm can have little plant cover at ground level so more than 20% of the ground could be leaf litter. However, a plot with 80% coverage of short, dense grass could have no more than 20% leaf litter cover.
- % DOWNED LOGS (GCLOG)
- percent of ground covered by downed logs (logs >12 cm diameter). This value should be independent of taller, sparser vegetation (can sum to more than 100%), but dependent on low dense vegetation (sum to 100% or less). SEE LEAF LITTER EXAMPLE
- % ROCK COVER (GCROCK)
- percentage of ground covered by rocks (mineral substrate, the pieces of which are larger than grapefruit sized). The distinction between GCROCK and GCBARE is that nests could potentially be hidden among rocks, but would have to be placed on the surface of bare ground.
- % BARE GROUND (GCBARE)
- percent open ground not covered by leaf litter. This value should be independent of taller, sparser vegetation, but dependent on low dense vegetation. SEE LEAF LITTER EXAMPLE
- % WATER COVER (GCWATER)
- percent of ground covered by standing water. This value should be independent of taller, sparser vegetation, but dependent on low dense vegetation. SEE LEAF LITTER EXAMPLE
- % MARSH VEGETATION (GCMARSH)
- percentage of ground covered by marsh vegetation (vegetation undifferentiated by species or type that is growing in water). If sedges are not reported separate of other marsh vegetation (in GCSEDGE), then leave GCSEDGE blank instead of reporting a 0 cover for sedge.
Measurements Taken Within Larger (11.3 Meter) Vegetation Plots
The 11.3 meter radius plots are used to count the numbers of trees near the center of the vegetation plot. Live trees are separated into the size classes given in the table, below. Separate counts should be made for each species of tree in the plot. Snags (dead trees) taller than 1.4 m and >12 cm DBH should be combined in a single category (don't separate snags by species for BBIRD data). Some cavity-nesting species will nest in trees smaller than 12 cm DBH. If you have species nesting in snags of <12 cm DBH then don't ignore small snags. Create an additional size class for snags smaller than 12 cm DBH. If for your own work you want to separate snags into a greater number of size classes, please make 12cm dbh the boundary between two of your size classes so that we can easily combine the data in the BBIRD database.
TABLE: TREE SIZE CLASSES
Live Trees (measure each species separately) Small trees: 8 -- 23 cm dbh Medium trees: 23 -- 38 cm dbh
Large trees: >38 cm dbh
Snags (combine all species) Small snags: < 12 cm dbh and > 1.4 m tall Medium snags: > 12 cm dbf and > 1.4 m tall
The following are the variables recorded for each of the large diameter plots:
- RADIUS, LARGE PLOT (TSRADM)
- radius (in meters) of the plot used to count numbers of tree stems. This will almost exclusively be 11.3 m
- TREE SPECIES (TREESP#)
- species code for the #th species of tree encountered on the large vegetation plot. There is no specific order in which tree species must be presented
- NUMBER OF SMALL TREE STEMS (TSTM#CL1)
- number of live stems of <8 cm DBH, of species # within the large vegetation plot
- NUMBER OF SMALL/MEDIUM STEMS (TSTM#CL2)
- number of live stems of 8 -- 23 cm DBH, of species # within the large vegetation plot
- NUMBER OF MEDIUM/LARGE STEMS (TSTM#CL3)
- number of live stems of 23 -- 38 cm DBH, of species # within the large vegetation plot
- NUMBER OF LARGE STEMS (TSTM#CL4)
- number of live stems of >38 cm DBH, of species # within the large vegetation plot
Use as many TREESP#, TSTM#CL1, TSTM#CL2, TSTM#CL3, and TSTM#CL4 as required for all species of trees in the plot. Rare species can be pooled into the category "OTHER".
- NUMBER OF SMALL SNAGS (M11SNAG1) number of snags (all species combined) less that 12 cm dbh
- NUMBER OF LARGE SNAGS (M11SNAG2) number of snags (all species combined) greater that 12 cm dbh
Field Protocols
Last modified: Sunday, 23-March-97 12:45:30 MST