United States Department of Agriculture
Forest Service
Pacific Southwest Research Station

General Technical Report

Other Considerations

Color Banding Individuals

Observer variability can be a great problem in many of the censusing schemes described above (Verner and Milne 1989). The color banding of individuals allows field identification and survival estimates of individuals without recapture and can greatly enhance spot mapping efficiency, the ability to find nests, and basic life history information. Furthermore, it allows more detailed observation of behavior including breeding biology, survival, and foraging ecology. Color-banding and other auxiliary markers must be authorized by the Bird Banding Laboratory.

Methods of Habitat Assessment

Many applications of habitat analysis are in the literature (e.g., Verner and others 1986). It is not our intention to outline what analyses can be done, but to emphasize that, at the least, vegetation information should be taken at each of the stations. Objectives of vegetation assessment can be many, but among the most common are to relate, in one way or another, the changes in bird composition and abundance to differences in vegetation. These vegetation changes can be either changes over time, or differences between habitats. Two adequate, but relatively time-consuming, methods of habitat assessment are those of James and Shugart (1970), used primarily in forested habitats, and Noon (1981). An excellent and rapid method which could be substituted for the method of estimating stand characteristics below is that of MacArthur and MacArthur (1961) which involves estimating foliage density. The technique uses horizonal measurements to estimate density by relating the percentage of a board that is obscured by foliage. This method has been tested and found reliable by Conner and O’Halloran (1986) and Conner (1990).

  If managers wish to characterize the interactions of birds and habitat in a region, then some kind of habitat classification with sampling in proportion to the relative abundance of habitat in that region is the optimal design. This sampling, stratified by habitat, should be done with the guidance of a biometrician.

  We present two alternatives here. One is that used to type vegetation into broad habitat classifications, as the Constant Efforts Site vegetation assessment technique does, or a more specific one, involving estimation of stand characteristics. We strongly suggest the latter method, as being more useful for monitoring.

Broad Habitat Classification

Objectives—This method provides brief, overall classification of vegetation and a map that allows other investigators to evaluate the habitat of your station. These data should be the minimum collected on vegetation at any monitoring station. If more detailed vegetation data are collected, then this level need not be taken.

  Considerations—The information collected should provide enough data to determine the vegetation types. The method will not provide quantitative information for correlative analyses and ordinations.

  Procedures—It is best to make a map of the main areas of habitat within the station on a yearly basis, sometime in June. Prepare it on the scale of approximately 1:2000 (approximately 1 foot to a half mile [1 m to 2 km]). Include the major vegetation types, extending it at least 100 meters beyond the outermost net or capture location. Indicate on the map: trails, roads, ditches, streams, marshy areas, net or census points, open water, and broad habitat boundaries. Also on the map should be a reference point identifiable on a U.S. Geological Survey topographic map or equivalent.

  Use colored lines to separate habitat types, and record the following on a form:

  • Habitat type: broad category such as forest, brush, marsh, field, etc.
  • Shrubby vegetation: list the shrub species comprising more than 10 percent cover in order of their percent cover in each type.
  • Trees: list the tree species comprising more than 10 percent cover in order of their percent cover in each type.
  • Height of vegetation: record the approximate average height of the canopy of forest or brush to the nearest meter.
  • Ground layer: describe the vegetation of the ground layer in terms of the common name of the main species groups present, e.g., ungrazed grass, bare ground with nettle, rushes, etc.
  • For wet areas: indicate the water depth in June, or for temporary ponds and streams, give the period that water was present.

Estimation of Stand Characteristics

  Objectives—This is a system for assessing habitat characteristics in an efficient and timely fashion at vertebrate monitoring stations. It is taken from a method developed by Bruce Bingham and C.J. Ralph.

  Considerations—The information collected will provide enough data to determine the vegetation formation, association, and major structural characteristics. The types of data are those which have some logical relationship with bird requirements for feeding or nesting. The method provides enough quantitative information for correlative analyses and ordinations. It is flexible so that it can be applied to any vegetation formation, including deserts, grasslands, and forests.

  Procedures—Establish a releve, a variable radius plot centered, for example, on a census point. The size of the plot will vary, depending on the homogeneity of the vegetation composition, and the density of the vegetation. Generally, this would be a radius of less than 50 m, and often about 25 m. Walk around the point for no more than 5-10 minutes, or until you stop adding new species, whichever is less. Once the search is stopped, the distance from the stopping point, or the outermost boundary of vegetation that the observer can see from the point center, is the radius of the plot and is treated as a boundary for estimating relative abundance.

  If the point has more than one vegetation type, then establish two releves. An example would be along a road, with a clear cut on one side, and a mature forest on the other. No more than two releves should be established at a point.

  Determine the number of major layers of vegetation within your releve by their dominant growth form: tree layer (T), shrub layer (S), herb (H), and the ground cover (moss and lichen) layer (G).

  In a forest with all layers, the tree layer is the uppermost stratum, dominated by mature trees. It may be a single layer, or consist of two or more sublayers recognizable by changes in density and canopy status (see below). The shrub layer is dominated by shrubs or small trees. The herb layer is dominated by low-growing plants, typically nonwoody, although seedlings and other reproduction of trees and shrubs may be present. The ground layer is dominated by such plants as mosses, lichens, and liverworts. Bare ground and litter are ignored for this classification scheme.

  We recommend the use of the following height classes for each stratum, if they are appropriate, because they can make the process less subjective. For example, the tree layer could include any plants taller than 5 m (In shorter forests, this might be lowered to 3 or 4 m, as appropriate). The shrub layer could then be established at between 50 cm to 5 m. The herb layer includes any plants less than 50 cm tall. The moss/lichen layer refers to a ground-appressed, low carpet, less than 10 cm high.

  For purposes of bird-habitat association, only species of trees and shrubs need be identified and recorded in the data below. For other plants, a common name such as FERN, HERB, MOSS, or LICH will suffice for most purposes. Plant ecologists have used some species in the herb or ground cover layers as indicative of a particular plant association. In this case, the species should be recorded.

  Determine the average height of each major layer present and dominant plant species. It is desirable to have additional information on structure, such as the maximum and minimum d.b.h. of canopy trees and total percent cover value of each layer.

  Determine relative importance of species in each layer present. Importance can be expressed as either abundance or cover. Percent cover is probably the most common, and we suggest using it.

  Below is a detailed description of the data we suggest be taken and recorded as on figure 15. The data are separated into Location Data and Vegetation Data.

Location Data:

State or province—The 2-column code for each.

Region—An 8-column code, designated by the investigator. Often, the name of the USGS quad, a prominent landmark, or a nearby town will provide the best code name.

Site data.

  • Latitude and longitude—For each point, latitude and longitude should be recorded as the southeast corner of the 1-minute block containing the point, as determined from accurate topographic maps. For example, 40°53’20"N, 124°08’45"W would be reduced to 4053-12408.
  • Elevation to nearest meter, by using an altimeter.
  • Aspect of the slope (the compass direction the observer faces when looking down hill) to the nearest degree, with a compass.
  • Percent slope, with a clinometer.
  • Presence (+) or absence (-) of water within the releve.
  • Plot radius, distance from the center to the edge of the releve.

Vegetation Data:

Vegetation Structure and Composition

  • Total cover—Estimate the cover of each of the four layers, according to the established scale such as Braun-Blanquet (Mueller-Dombois and Ellenberg 1974) or Daubenmire (1968). We recommend the Braun-Blanquet Cover Abundance Scale, which is: 5 => 75 percent cover; 4 = 50-75 percent cover; 3 = 25-50 percent cover; 2 = 5-25 percent cover; 1 = numerous, but less than 5 percent cover, or scattered, with cover up to 5 percent; + = few, with small cover; and r = rare, solitary, with small cover.
  • Height—Record to the nearest decimeter (0.1 m) the average height of the lower and upper bounds of each of the four layers.
  • Species—Record the species by a 4-letter code (using the first two letters of the genus and the first two of the species) with the greatest cover (foliage or crown cover) within each layer’s boundary.
  • D.b.h.—For each layer where trees are present, record the diameter at breast height to the nearest centimeter of the largest tree in the layer and also for the smallest trees.
  • Species—Record the species of trees used for minimum and maximum d.b.h. measurements
  • Number of sublayers—Sublayers are useful to give the plant ecologist a quick overview of the structure of a layer, and are primarily relevant to the tree layer, although sometimes seen in the shrub layer. Record the number of sublayers visible in each primary layer. Record “1” if the layer is uniform and “2” or more if more than a single layer is divided into sublayers. In a primary layer, sublayers are sometimes obvious because of one or more species with shorter heights than the dominant species of the upper portion of the layer. In addition, sublayers are sometimes formed by cohorts of one or more size classes, possibly related to some event. For example, the tallest trees in a stand may form an open (low-density) layer of emergent individuals. Beneath that may be a denser layer of trees forming the main body of the tree layer. Below this denser layer may be another open or closed layer of trees that are intermediate to the main body of the canopy. This layer may consist of shade tolerant species or reproduction. Biologists should be cautioned that extreme precision is not required for this estimation, and unless sublayers are very obvious, they should not be recorded.

Species composition data

  • Sublayer—For layers where sublayers have been recognized, record the sublayers with a letter designating the primary layer, followed by a number (e.g., T1, T2, T3, S1, etc.), indicating the sublayers by decreasing heights.
  • Cover or cover abundance value, as above, using the Braun-Blanquet method—Because of the difficulty of determining crown covers independently for species of trees in a canopy, sometimes basal area cover of stems (trunks) has been used for tree layer species and crown cover for species in other layers. We suggest the cover abundance value for consistency.
  • Species—Record the species’ name for each plant species making up at least 10 percent of the cover.

  Additional/optional information can be integrated into the method, if desired:

  Snags: list layers with snags present; separate into those with a d.b.h. of less than 10 cm and those larger.

  Logs: list those less than 10 cm diameter at large end by abundance or cover class, and those greater than 10 cm.

  Comments—This type of vegetation assessment is limited by the size of the plot and the amount of estimation required. For example, a plot of even 50 m in radius obviously does not include all vegetation inhabited by birds heard or seen from a census point. This would require a plot of 200 m or more radius. However, most birds detected at a point are within 100 m, and many are within 50 m. Further, time limitations would require much more time spent monitoring vegetation than spent counting birds.

  When observers are required to estimate, a substantial amount of error is introduced. What effect the degree of observer error likely with estimation would have on conclusions should be established. The principal source of error in this method of vegetation assessment is the determination of percent cover and heights. Intensive training can moderate this source of error, enabling each vegetation assessment to be placed into at least broad categories or plant associations.

Weather Monitoring

We suggest the following weather measurements three times per day, at the beginning, midpoint, and end of the census or capture period in a day. The maximum high and low temperature from each 24-hr period should also be recorded. Data from nearby weather stations may also be used. However, some measurements from the immediate area are more valuable.

  • Weather—Use RAIN, DRIZZLE, SLEET, SNOW, or FOG. (If it is raining and foggy, put down RAIN.) If the above conditions do not apply, use: OVC (overcast), more than 90 percent cloud cover over entire sky; BRK (broken), 50-90 percent cloud cover; SCT (scattered), 10-50 percent cloud cover; or CLR (clear), less than 10 percent cloud cover.
  • Wind Direction—Using an anemometer, stand facing into the wind and record the direction to the nearest 1/16th of the compass, i.e., N, NNE, NE, etc. If winds are variable, record predominate direction.
  • Wind Speed—Record both the average and maximum speeds.
  • Visibility—Estimate visibility to the nearest 250 m if less than 2 km, otherwise to the nearest kilometer.
  • Barometric pressure.
  • Temperature—Dry bulb temperature. Record to the nearest 1 degree centigrade.
  • Relative humidity.
  • Rain—Record from a rain gauge to nearest 0.1 mm.

  We suggest a continuous strip chart recorder to measure temperature and a somewhat permanent station to measure rainfall.

Publishing Information

Front Matter

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 6