In 1994, the Northern Region of the USDA Forest Service (USFS) initiated a region-wide landbird monitoring program so that managers might better understand the habitat relationships of landbirds that breed in the northern Rocky Mountains and, in the future, might be able to assess longer-term landbird population trends. The program was initiated to help the USFS meet its legal mandate (National Forest Management Act of 1976) to monitor populations of "indicator" species as a mechanism to maintain viable populations of native vertebrates. Landbirds are a good indicator species "survey group" (Hutto 1998) because they are highly visible and many species can be surveyed simultaneously. Combining data from several Forests permits an assessment of trends and habitat relationships over the larger Region and provides an indication of changes in relation to land management practices.
The philosophy of the LBMP has been to produce short-term results on habitat relationships and management questions as well as long-term trend monitoring. Products already published include a quantitative description of the distributions of bird species across the major vegetation types in the Northern Region (Hutto and Young 1999), regression models to expose finer-resolution patterns of occurrence due to the continuously variable nature of vegetation features within cover types (Young and Hutto 2002), and an exploration of landscape level variables affecting the distribution of Brown-headed Cowbirds throughout the region (Young and Hutto 1999). The LBMP program not only collects habitat data with permanent monitoring points so that correlations with land use and cover type can be determined, but also conducts short-term studies of the effects of specific management practices on selected bird species.
The sampling design for monitoring consists of a series of almost 400 survey transects, each consisting of 10 points that are permanently marked and sampled once each year. The distribution of points is geographically stratified by 7.5-minute topographic quad maps and incidentally stratified by vegetation cover type. The placement of long-term points along roads greatly increases our ability to relocate permanently marked points, reduces travel time between points, reduces the chance of injury to field workers, and still allows one to detect all species known to occur within a region (Hutto et al. 1995). These transects are visited every two years, leaving the alternate years for more focused assessment studies. In 2002, we will be returning to our permanently marked, long-term monitoring transects.
LOCATIONS OF LONG-TERM MONITORING TRANSECTS
Permanent monitoring transects were established on each national forest in the Northern Region in 1994. Point transects were initially mapped by forest and district biologists, and individual points were later marked in the field by the bird technicians.Transects were placed along secondary or tertiary, open or closed roads or trails, but not on paved roads, highways, or major thoroughfares. We provided for a relatively even geographic distribution of transects across the roaded areas of each national forest. The target area included lands not designated as wilderness (logistically infeasible) and any topo quad that contained at least 25% Forest Service lands; within these constraints, no particular project areas were targeted to the exclusion of other areas.
Generally, 1 to 4 transects were mapped in each topographic quad. We determined the needed number of transects per quad by using the following formula: [(number of field workers) x (number of field days available for conducting counts) / (number of topo maps within the set)]. For example, if the target area was completely covered by 20 quads, and there were two observers available for conducting counts for 30 days each, we assigned 3 transects per quad. It is worthwhile to generate an additional transect on each quad to serve as a "buffer" in case one of the other transects proves to be inaccessible. If only a portion of the Forest occurred within a given quad, we reduced the number of transects for that quad accordingly. The idea was to spread the transect sites broadly across the Forest.For logistical reasons, it was necessary to have observers travel as teams, which further constrained the exact distribution of transects among quads.
The placement of start points for transects was random after stratification by USGS topographic quad. Each quad map was divided into equal areas. Using a random-number table, two 2-digit numbers were chosen to represent coordinates within an area and a point was positioned using these coordinates. This was repeated for each area within the quad map. From each random point, the nearest road or trail was located to identify the starting point for a transect route. The starting point for the transect was that place on the road or trail nearest the random point. However, many start points were moved to the nearest intersection or other landmark to make them easier to locate. That point was marked on the topo quad and the appropriate aerial photo. Acceptable trails included those that were maintained, and acceptable roads were unpaved secondary or tertiary roads (preferably the smaller one-lane roads typical of those created for management purposes). In the absence of a road or trail, that area of the quad was discarded. The forest travel map and District personnel were consulted for ease of access and seasonal or permanent road closures. We rejected areas more than 1 hour walking time from the nearest parking place or more than 1 to 1.5 hours driving time from the nearest camping place. Permission was obtained for travel behind some locked gates for a day to obtain the census data. Otherwise, an alternate transect was selected in the quad. The direction of travel from the starting point was sometimes dictated by the only direction of travel possible. Otherwise, direction was determined by the flip of a coin. Total transect length is about 1.5 miles.
We made a set of neat 8.5" x 11" photocopies of the mapped transects for use in the field (including a map edge is useful for orientation), highlighting the location of the transect. A color photocopy of a recent aerial photo of the area was also made. In permanent black ink, we labeled the top edge of each topo map copy and aerial photo with the district name, topo quad name, and transect name. The exact location of individual points was left to be done in the field.
A Transect Location Form (Appendix II) was completed with as much information as possible for each transect--in particular, the directions and access information. The transect was named (usually with a local geographical feature). Each transect also was given a nine-digit number equal to the latitude-longitude of the start point to the nearest minute (e.g. 464511404). This number was unique for each transect and was an important identifier in the database.
It was very important to have an "index" to the quad and transect locations on a National Forest travel map, for use both in the field and by the data management center. We used a black marking pen to outline and label (with the most recent USGS quad name) the topo quad locations.(Note that most Forest maps indicate quad corners with a faint cross mark. These can be used to quickly grid in the quads with a ruler and black marking pen.) With a highlighter, we indicated the location of each transect within each quad and labeled each transect with the name or number. One copy was provided to each crew member. A set should also be made for permanent record at each National Forest S.O.
Sets of topo map, aerial photo, and transect-location-forms for each transect were organized in a 3-ring binder or folder for each National Forest, ordered by district and/or topo quad. Plastic top-loading sheet protectors for 3-ring binders were extremely useful for long-term storage, protection, and organization of the forms, maps, and aerial copies. The Forest travel maps were included as an overall index of transect locations.
POINT MARKINGS AND DOCUMENTATION
The precise placement of count points along a transect route were determined by a field technician in the first year the transect was established. The points were not preferentially located in habitats or management treatments, but were positioned at 300-meter intervals as determined by precise pacing. Where there were curves or switch-backs, points were placed at least 250-meters straight-line distance apart.
Points were marked in the field with a round numbered aluminum tag (1-10) either on a wooden stake, a rebar stake, or on a tree, snag, or stump. Trees and stumps were used preferentially wherever these were close to the roadside; otherwise stakes were used. Aluminum nails must be used in trees because these will not harm chainsaws or personnel if the tree is later harvested. At least a half inch of nail was left showing on live trees to allow for growth. On trees, markers were placed at eye-level facing the observer as a field worker proceeds along the transect.
The documentation for each transect includes a written description on the back side of the transect location form of the general surroundings and placement of each marker. Crews should be as informative as possible in writing field descriptions of directions to the transect and location of the points on the transect location form. An accurate description is obviously helpful in relocating points. Recording "point #2 in Doug-fir on right" may not offer enough detail depending on the habitat. Additional information, such as "point #2 in 35-cm dbh Doug-fir on right of trail, next to broken-top snag, 20 m before sharp left bend in trail" is much more helpful. Field workers may add information to the description in subsequent years if they feel it is warranted.
Field observers mapped point locations onto topos and aerial photos while in the field. The maps aid tremendously in relocating points in subsequent years and facilitate locating points on a GIS database. This is easier to do on the aerial photos because more landmarks are available. The aerial photos, therefore, had the most precise estimates of point locations and were used by personnel in the GIS lab to position points as precisely as possible. Because each point location is entered into a GIS database, it is critical that the observer be as painstaking as possible in placing point locations on the aerial photo. Even if a trail or point is not easily visible on the aerial, the observer should carefully estimate positions. The point locations are positioned in the GIS exactly as marked on the aerial photo. If the aerial photo is current, it will match the GIS satellite image fairly closely and points can be pinpointed using the landscape patterns. GPS readings at points will also be helpful, of course, although these may not be possible in dense forest locations.
After the initial field season, the field crews take copies of the marked aerial photos into the field each time to aid in locating the points, and to double-check that point locations are placed correctly. If documentation for existing transects is incomplete or inaccurate, observers fill in any missing information and make corrections. Any such revisions are noted on the Revisions Form provided (see Appendix III). Inadequate documentation or the intentionally cryptic point markers may cause delays in relocating existing points in the field. If after a reasonable time searching (e.g., 10 to 15 minutes) a field worker cannot locate a point marker, (s)he places a new point at the appropriate distance from the last point and as close as possible to where the point was originally mapped. Again, any changes or additions are noted on the form provided (see Appendix III).
Before traveling to the transects every year, the forest travel map and district personnel should be consulted for ease of access and seasonal or permanent road closures. Most transects that required crossing private property to access were rejected. In the rare exceptions where this is necessary, generally permission must be obtained every year.
LOGISTICS:
The logistics associated with this field effort are not a trivial concern. Crews camp in the vicinity of the transects and move camp every 1 to 2 nights. Work centers, trailers, or cabins are sometimes used for nearby transects, rain days, or weekend housing. However, the use of such bases, while more comfortable, is avoided if total pre-dawn travel time to a start point would exceed 1 to 1.5 hours. Camping is preferred to minimize travel time. A balance is struck between the field crew's reasonable comfort and time for travel, data collection, data entry, logistics and organization. For safety reasons, field radios are provided. If crew members camp separately, they should arrange a schedule of checking in with their contact in the local district or forest office.
Each observer conducts one transect per day independently. An equal number of routes is assigned to each observer, making sure that the geographic and elevational distributions of routes are as similar as possible among observers. The order of visits to transects is set by elevation and seasonal access. Relatively low-elevation transects are visited first, relatively high-elevation sites last. (Both accessibility and life history stage of the birds are most similar among points if the higher elevation transects are left until later). This may require more than one visit to an area during the season because not all nearby transects may be accessible at the same time.Transect visits are scheduled as closely as possible to the date run in previous years.
SEASON:
In the Northern Region, counts are conducted from the third week of May through the second week of July, depending on local bird phenology. Many grassland species are well into their breeding season in May and finish by late June. Highland species, on the other hand, start a little later, and the third week of May is usually the earliest date that counts should be conducted. Throughout most of the Region, family groups are moving and the breeding season is over by the second week of July.
START AND FINISH OF A TRANSECT:
Field observers are in position to begin their first point count of the day by 15 min after sunrise, which is usually sometime between 0615 and 0645, Mountain Standard Time (Montana and North Dakota), or 0515 and 0545 Pacific Standard Time (Idaho). Thus, counts begin after the pre-dawn chorus and continue throughout the period during which bird activity and song is relatively constant. Point counts are completed by 1100 Mountain time, or 1000 Pacific time. This period allows ample time to complete 10 point counts, walking between points.
UNACCEPTABLE FIELD CONDITIONS:
Don't bring your dog or friends along; they may influence the results in a manner that cannot be adjusted later. Don't bother collecting data when the weather is bad enough to influence bird activity--that includes continuous rain (but not light drizzle) and wind that is constant and of enough strength to bend the tops of trees (Beaufort 5).
Wind on the east slope of the Rockies and in the Little Missouri National Grasslands is constant enough that it may affect observations, but birds are generally still active. Therefore, proceed with counts when strong winds are the norm. Unacceptable conditions in these areas would include brewing storms, wind coupled with steady precipitation, etc.
The weather across the Northern Region is so variable that field observers should travel to the transect start point before assessing weather conditions. Oftentimes, a pre-dawn rainstorm will let up soon after first light, or wind may pick up at dawn and then settle again within 15 or 30 minutes. In other words, weather conditions at a transect site cannot be assessed from inside a warm sleeping bag miles away.
If you find yourself sitting in the truck at 6:00 am wondering about the steady drizzle--wait. Rule of thumb is to wait up to 1 to 1.5 hours, or until you cannot reasonably complete the transect by 1100 hours.
Most importantly, under all circumstances safety comes first. If weather (e.g., lightning, cold, rain, snow) or road conditions are placing you at risk, STOP. Take cover, get warm, pull over, or do whatever you need to do to get yourself to safety.
The data booklets are designed to contain all data for one transect per booklet. General transect information is entered on the cover, and the bird and habitat data for each point are entered inside on facing pages.Always fill in every variable on the field forms. Blanks are interpreted as missing data, NOT as zeros.Likewise, a zero is NOT to be used to represent missing data. Field forms with missing information compromise the reliability of your data.
It is absolutely essential to double-check your entries on data forms and maps on the same day that each transect is run. Casual oversights by observers such as recording incorrect cover type and species codes, incorrect or abbreviated quad names, leaving blanks where zeros were intended, and not marking or incorrectly marking topos and aerials, create enormous headaches for us later. We run computer routines to uncover such errors, so we know who submits reliable and who submits unreliable information, and we use this information when it comes to job recommendations later.
The formats for entering data in each section of the field data booklet are presented in Tables 1-4 below:
Table 1. Instructions for recording information on the cover of the field data form.
VARIABLE |
EXPLANATION |
DATE |
Use 2 columns for month, 2 for day, and 4 columns for year, for the date on which the bird count was conducted; e.g., 0617 2000 = |
REGION |
Forest Service Northern Region always = 1. |
2-digit code (Appendix V) and name written out. |
|
DISTRICT |
2-digit code (Appendix V) and name written out. |
STATE |
Two-letter state codes; e.g., ID, MT. |
QUAD |
Exact name of topographic quad as it appears on the transect-location form (including spaces). |
TRANSECT NAME |
Unique name given the transect as it appears on the transect -location form. (NOT entered in the database) |
TRANSECT NUMBER |
Unique 9-digit number (4 digits for latitude and 5 for longitude) as it appears on the transect-location form. |
In a point count survey we record all birds that we detect by any means during the 10-minute time period. We may detect and identify a bird using song, call, or visual cues. It is important to get as complete a record as possible of the birds present on the site during that period. But we know we are not detecting all the birds that are actually present. We may even have seen or heard other birds before or after the 10-minute period. We do not record these because we need to get a consistent, quantitative index of abundance at each point that we can compare across sites and across years.
The technique follows recommendations established by participants in the national point count workshop (C. J. Ralph et al. 1995). A description of the point count method can also be found in Hutto et al. (1986). In general, a 10-minute point count is conducted at each of the 10 sampling points that constitute a road- or trail-side transect. Points are visited once during the breeding season between mid-May and early July. All birds seen or heard within the count period are recorded. "Spishing" is not allowed during a count, but is a perfectly acceptable way to attract nearby birds after a count to get positive identifications (just be sure you don't attract and record some bird that wasn't detected during the count!).
We use 10-minute rather than 5-minute counts for two reasons: (1) 5-min counts may simply be too hurried to ensure reliable and consistent data among observers, i.e., there is probably considerably greater inter-observer variation in 5-min than in 10-min counts; (2) data from 10-min counts are compatible with the data we have already gathered for a habitat association studies, and this allows us to use results from the from the permanent transects to build and/or test our habitat-relationships models.
Top of form: At each point, observers record their name, the date, stop number, time-of-day at the start of the count, wind, sky, temperature and stream noise conditions at the top of the form.It is imperative that this information be recorded on the data form for each and every point because we use this information to organize the hard copies, which we need to consult when questions arise about specific data.
Table 2. Instructions for completing the top of the bird form.
VARIABLE |
EXPLANATION |
OBSERVER |
Use observer's first 2 initials and full last name. |
DATE |
Use 2 columns for the month and 2 for the day on which the bird count was conducted; e.g., 0617 = June 17. (Do not use special date types in database.) |
YEAR |
4-digit year (Y2K compliant!) |
STOP |
Stop (point) number, should always run from 1 to 10 |
TIME |
Use the 4-digit military time-of-day the count is started at point; e.g., 0814. |
WIND |
Use the Beaufort wind scale codes (0-5) as provided in the field booklet. |
SKY |
Use the codes (0-6) defined in the field booklet and Appendix V. |
TEMP |
Use a thermometer to record air temperature at the point to the nearest degree (Fahrenheit). |
STREAM |
Use the codes (0-4) defined in the field booklet for description of stream noise and its probable effect on bird detectability. |
Recording bird detectionsOne row in the data booklet is used for each detection of a bird (or pair or group of birds at the same location; the number recorded under "Abundance"). Therefore, there can be more than one entry for a given species on a point count. We use standard 4-letter bird codes from the national Bird Banding Lab (with a few exceptions). These are listed in Appendix VI and in the data booklets and must be used exactly as given. Some codes are very similar and an error can be serious. The codes are based on the first two letters of the first and second parts of the common name of the bird species (or the first letter of each part of a hyphenated portion). You should write out the name of any bird species for which a 4-letter code is uncertain, otherwise you'll end up with a nonsense code, and you won't remember which bird species it was supposed to represent.
Table 3. Instructions for recording bird detection data.
VARIABLE |
EXPLANATION |
SPECIES |
Record the appropriate four-letter bird species code for a given detection. Flyovers should be included as well. A master list is included in Appendix VI and in the field booklet. You should double-check codes. If you are not sure of the identity of a species, then put a question mark next to the line and exclude the detection if there is no subsequent verification. |
ABUN |
Estimate the number of individuals detected at the associated distance and bearing. Should always be 1 or greater - no blanks. |
DISTANCE |
Estimate distance to the bird(s) to the nearest 5 m inside 50 m, and to nearest 10 m beyond. Flyovers should be coded as "999". |
CUES |
How the bird was identified; both columns must be completed independently: A = Audible (0 = not heard, S = Song, D = Drum, C = Call); V = Visual (0 = not seen, V = identified by sight) |
LOCATION |
Record whether the bird was detected within or outside the cover type at the point. 0 = inside the main cover type; 1 = inside "Edge 1" habitat; 2 = inside "Edge 2" habitat; 9 = over 100 m away or flyover. |
Distance: Estimate the horizontal distance to every bird detection, to the nearest 5 m inside 50 m, and to nearest 10 m beyond. Flyovers should be coded as "999".
Distance estimation is very difficult and requires a great deal of practice. First we have to be able to estimate distance well to a visual landmark and that in itself takes a lot of practice. It also takes an accurate pacing ability in varying topography. These skills are practised during the training session. Then comes the hard part: estimating how far away a SOUND is. We may do it directly, by how loud it sounds, or we may look to see where it sounds like it is coming from and then visually estimate the distance to the physical location (vegetation) from where we think the sound came. It is not clear which is best because we could be easily fooled either way. It helps to walk out and find a singing bird (actually many, many birds) after you have estimated the distance. But remember that the next bird of that species you hear that loud may NOT be at the same distance. We really need to track down enough individuals to have a feel for the fact that loudness can vary depending on species, vegetation, topography, what direction the bird is facing, and perhaps even enthusiasm!
When estimate distances, please do NOT assume that we will be cutting off the data at any particular point. Just because we sometimes analyze data based on the birds within 100 m does NOT mean that you should bias your estimates to include all of the borderline birds within that radius. We often analyze the data in other ways, and if we try to calculate detectability profiles based on biased distance estimates we will get garbage.
Cues: The consistency of the "index of relative abundance" that we are recording depends on the many things that affect the detection probability of the birds. Detection probability is affected by such things as the behavior of the bird, the density of the vegetation, and the attentiveness of the observer. These variables have different effects on the different cues we use to detect and identify birds. For example, the ability to detect a bird by sight is more strongly influenced by vegetation then is sound detection. When we analyze the bird data from our monitoring program, it is very helpful if we know which birds were detected by songs and which by sight. We will be keeping records of this in our field data this year.
We want to know the cues that were used to detect and identify each bird. If you heard and were able to identify the bird by sound, mark that in the first column. Otherwise write "zero". If you saw the bird and were able to identify it visually, write "V" in the second column. Otherwise write "zero". The two columns are independent; either or both can be non-zero. A sound detection in the first column can be one of three categories. If you hear the primary "song" of a passerine "songbird," use "S". If you hear a woodpecker or grouse "drum", snipe "winnow", or nighthawk "boom," or any write "D." If you did not hear one of these but instead heard and identified the bird by some type of "call," write "C." This includes all vocal sounds of birds that do not have a recognizable "song" (e.g., swallows, corvids, nuthatches, and non-passerines), or calls of passerines other than the primary song (e.g., WETA "piterik", TOSO call note, warbler chip, etc.).
It is important to check the accuracy of your data records after each point, BEFORE going to the next point. Make sure that all necessary boxes are filled in (NO BLANKS), and that all of the codes are correct. People do write the incorrect codes on occasion, e.g., they see a House Finch, but write HOSP by accident; or they consistently use a wrong code, such as MTCH instead of MOCH for Mountain Chickadee. It's easy to catch such errors when you review your entries immediately, but it's hard to remember what you saw a day or two (or even an hour) later.
If you fail to finish a point count (some form of disturbance or weather forces you to quit), do NOT enter those data into the computer; rather, note the time spent at the point in the comments section at the bottom of the bird data page. Make it clear that the count was not completed, and draw a line through the page. If you conduct a legitimate 10-minute count and don't get any birds, that's fine; enter "NONE" under species and leave the rest blank.
General measures of vegetation structure are recorded at each point on the return walk along the transect. The vegetation data are being collected for two reasons: (1) such detail is unavailable through remote sensing but may be useful in terms of classifying each point according to elements that are important to bird species and, thus, in explaining any observed changes in density over time; (2) measurements provide additional data that will help with the verification of vegetation cover types that are being mapped by the University of Montana Wildlife GIS Lab. In subsequent years, observers will continue recording vegetation measurements in order to provide repeated sampling and to enable comparison among observers.
The cover type classifications we use are based on a generalized scheme that is applicable throughout the northern Rocky Mountains (see Appendix VII). This cover type classification corresponds with the University of Montana Wildlife GIS Lab's cover classification based on satellite imagery.
GENERAL PROCEDURE:
At each point, ocular estimates are made of cover type and structural characteristics of the surrounding vegetation. It is important to remember that these are "quick-plots", intended to capture the gestalt of a broader area within which the point falls. These vegetation plots should rarely take more than 5 to 15 minutes to perform. Do not spend inordinate amounts of time trying to determine exact percent cover, for example, or identifying every plant species.
Because of the lack of vegetation on the road proper, it is best to move off the road in order to characterize the surrounding habitat. Observers walk into the stand to allow an accurate assessment of percent cover and plant species composition in the primary habitat. Otherwise, they may miss important characteristics of the vegetation structure or species composition. Even though measurement radii (15 m and 30 m) may be smaller than the radius within which birds were detected, the measures are meant to reflect conditions of the general area out to at least 50 m, and probably beyond.
The first set of vegetation measurements (Table 3) refer to the entire 100-m-diameter circle surrounding the bird point. Although seemingly redundant, it is imperative to record the observer, date, and stop on every vegetation page because booklets can be separated and pages mixed up.
Table 4. Instructions for recording habitat data
VARIABLE |
EXPLANATION |
OBSERVER |
Should be identical to that used on the bird data page for the same point; if the vegetation data were collected on some other day or by some other person, note those differences in the margin on the data form. |
DATE |
Should be identical to that used on the bird data page for the same point. If conducted on a different date, note in the margin. |
STOP |
Should be identical to that used on the bird data page for the same point. |
COVTYPE |
5-digit code for the vegetation cover type within which the count point is positioned (not including the road). If two or more cover types occur within 100 m, note the additional types(s) under "Edge" below. If the point is precisely on the edge between two or more types, use the one with the larger geographic area. A list of cover type codes is included here and the back of the data booklet. On the field data form, also write out the name of the cover type. |
EDGE1 |
The edge of a second distinct cover type occurs within 100 m of the point. Use a 5-digit cover type code to identify the additional type. Pay special attention to the presence of a narrow band of deciduous streamside riparian vegetation (505xx), marshes (50100), or wet meadows (50200). If none, code as "99999". |
EDGE2 |
The edge of a third distinct cover type occurs within 100 m of the point (close enough to have influenced the birds detected). Use a 5-digit cover type code to identify the additional type. Pay attention to the presence of a narrow band of deciduous streamside riparian vegetation (505xx), marshes (50100), or wet meadows (50200). If none, code as "99999". |
MOSAIC |
Indicate if point is within a mosaic of two cover types. Fill in the codes for the two cover types that make up the mosaic. |
ROAD |
Point is either on or within 100m of a road. Code as 0=no; 1=yes |
ROCK |
Presence of rocky outcropping, cliff, or talus slope within 100 m (close enough and large enough to influence the birds detected). Code as 0=no; 1=yes |
WATER |
Is running water present within 100 m? 0 = no, 1 = yes. |
Cover type is a 5-digit code representing the vegetation cover type within which the count point is positioned. Because we usually analyze bird-habitat relationships based on the birds detected within 100 m of the point, we also record all cover types within a 100-m-radius circle.
The cover type classifications we use are based on a generalized scheme that is applicable throughout the northern Rocky Mountains. The Rocky Mountain landscape is considered to consist of a matrix of land units, each of which is relatively homogeneous in vegetation structure and plant species composition, and differs from adjacent units according to the same criteria. Each sample point, therefore, falls within one of a range of vegetation cover types defined by a combination of the dominant plant species in the tallest vegetation layer and the vertical and horizontal vegetation structure.
The basic cover type framework is one that includes vegetation types dominated by one or more plant species. Open lands are usually dominated by grasses or sagebrush (Artemisia). Several riparian cover types are used, including marshes, shrubby streamside areas, willow flats, aspen stands or cottonwood bottomlands. Because conifer forest stands frequently undergo natural or human-induced disturbance that creates structurally different cover types, we define our conifer cover types based on three different criteria: tree species composition, successional stage, and, for the earlier stages, structure following disturbance (amount of canopy remaining).
HUMAN-DOMINATED AREAS
10100 - URBAN HUMAN DWELLING - High-density areas of human dwelling, usually dominated by non-native vegetation. Includes areas that appear gridded or "blocked" on maps, as well as high-density subdivisions.
10200 - RURAL HUMAN DWELLING - Ranchettes, country homes, cabins, etc. set in the midst of predominantly native vegetation.
20100 - IRRIGATED CROPLAND, PASTURE - Homogeneous areas of planted crops or pasture grasses, usually fenced and irrigated. There are often irrigation ditches or equipment associated with such land types.
20200 - DRY CROPLAND - Also homogeneous areas of planted crops or pasture grasses, often fenced but not irrigated. The physiognomic uniformity and lack of plant diversity associated with a planted cropland usually helps distinguish this type from native grassland. Pasture becomes "non-native grassland (30xxx)" when it becomes less "controlled."
Cover type classifications in grassland and sagebrush/shrubsteppe habitats have been developed to represent both the vegetative physical structure and the level of grazing/browsing pressure. Edaphic and other environmental conditions are better accounted for with the inclusion of both factors, e.g. low physical structure and no grazing, high physical structure with grazing. Factors providing evidence of grazing include cow presence, recent cow pies, trampling, bare ground and cattle trails. Bare ground and cattle trails should not be used as evidence of livestock grazing unless corroborative evidence exists.
GRASSLANDS
Grasslands vary in Idaho and Montana from the Palouse prairie of the western valleys (dominated by fescues and usually with many forbs), to the short grass prairies of central Montana (where blue grama is often dominant), to mixed-grass prairie where medium-statured grasses such as wheatgrasses and needlegrasses become more dominant. There is no true tall-grass prairie in Montana. Therefore, the "high-," medium-," and "low-structured" grasslands discussed below are categories relevant to the area (as indicated by height thresholds).
30101--high-structured (>35cm ht) and dense cover (>70%); ungrazed
30102--high-structured (>35cm ht) and medium cover (40-70%); ungrazed
30103--high-structured (>35cm ht) and sparse cover (<40%); ungrazed
30104--medium-structured (15-35cm ht) and dense cover (>70%); ungrazed
30105--medium-structured (15-35cm ht) and medium cover (40-70%); ungrazed
30106--medium-structured (15-35cm ht) and sparse cover (<40%); ungrazed
30107--low-structured (<15cm ht) and dense cover (>70%); ungrazed
30108--low-structured (<15cm ht) and medium cover (40-70%); ungrazed
30109--low-structured (<15cm ht) and sparse cover (<40%); ungrazed
30110--high-structured (>35cm ht) and dense cover (>70%); grazed
30111--high-structured (>35cm ht) and medium cover (40-70%); grazed
30112--high-structured (>35cm ht) and sparse cover (<40%); grazed
30113--medium-structured (15-35cm ht) and dense cover (>70%); grazed
30114--medium-structured (15-35cm ht) and medium cover (40-70%); grazed
30115--medium-structured (15-35cm ht) and sparse cover (<40%); grazed
30116--low-structured (<15cm ht) and dense cover (>70%); grazed
30117--low-structured (<15cm ht) and medium cover (40-70%); grazed
30118--low-structured (<15cm ht) and sparse cover (<40%); grazed
301xx - GRASSLAND - BIG BLUESTEM DOMINANT
302xx - GRASSLAND - BLUE GRAMA DOMINANT
303xx - GRASSLAND - BLUE-BUNCH WHEATGRASS DOMINANT
304xx - GRASSLAND - BUFFALOGRASS DOMINANT
305xx - GRASSLAND - CRESTED WHEATGRASS DOMINANT
306xx - GRASSLAND - GREEN NEEDLEGRASS DOMINANT
307xx - GRASSLAND - IDAHO FESCUE DOMINANT
308xx - GRASSLAND - INLAND SALTGRASS DOMINANT
309xx - GRASSLAND - KENTUCKY BLUEGRASS DOMINANT
310xx - GRASSLAND - LITTLE BLUESTEM DOMINANT
311xx - GRASSLAND - NEEDLE AND THREAD DOMINANT
312xx - GRASSLAND - PRAIRIE SANDREED DOMINANT
313xx - GRASSLAND - THICK-SPIKED WHEATGRASS DOMINANT
314xx - GRASSLAND - THREADLEAF SEDGE DOMINANT
315xx - GRASSLAND - WESTERN WHEATGRASS DOMINANT
316xx - GRASSLAND - UNIDENTIFIED DOMINANT SPECIES
317xx - GRASSLAND - OTHER DOMINANT SPECIES
SHRUB-STEPPE
Shrub-steppe refers to non-forested, upland (usually xeric) areas with a shrub component mixed with grasses. Sagebrush or other upland shrubs may be quite sparse, which is why we have the accompanying categories of “dense” and “sparse” cover. However, if juniper trees are present, that takes precedence and is classified as juniper woodland (401xx).
320xx - SAGEBRUSH, SHRUBSTEPPE - BIG SAGEBRUSH DOMINANT
32001--high-structured (>1.5m ht) and dense sage cover (>15%); ungrazed
32002--high-structured (>1.5m ht) and sparse sage cover (<15%); ungrazed
32003--medium-structured (0.5-1.5m ht) and dense sage cover (>15%); ungrazed
32004--medium-structured (0.5-1.5m ht) and sparse sage cover (<15%); ungrazed
32005--low-structured (< 0.5 m ht) and dense sage cover (>15%); ungrazed
32006--low-structured (< 0.5 m ht) and sparse sage cover (<15%); ungrazed
32007--high-structured (>1.5m ht) and dense sage cover (>15%); grazed
32008--high-structured (>1.5m ht) and sparse sage cover (<15%); grazed
32009--medium-structured (0.5-1.5m ht) and dense sage cover (>15%); grazed
32010--medium-structured (0.5-1.5m ht) and sparse sage cover (<15%); grazed
32011--low-structured (< 0.5 m ht) and dense sage cover (>15%); grazed
32012--low-structured (< 0.5 m ht) and sparse sage cover (<15%); grazed
321xx - SAGEBRUSH, SHRUBSTEPPE - LOW SAGEBRUSH DOMINANT
322xx - SAGEBRUSH, SHRUBSTEPPE - SILVER SAGEBRUSH DOMINANT
323xx - SALINE/ALKALINE SHRUBLANDS - GREASEWOOD DOMINANT
33000 - DRY SHRUBFIELD - Grasses generally fill area between shrubs. These must be upland shrub species (no riparian) other than sage or greasewood, and NOT a clearcut or other successional stage of a disturbed forested area (no tree stumps or seedlings).
34000 - MESIC SHRUBFIELD- Deciduous shrub field that has not been logged (similar in structure to clearcut tall shrub) and is not riparian. Often a result of natural disturbance (avalanche chute, blowdown area) but doesn’t have standing dead. Also not dry.
401xx - JUNIPER WOODLAND - Grasses and sage generally fill space between trees. Although junipers are conifers, they are short-statured trees that grow sparsely in open landscapes and are not subject to commercial logging. Therefore, we do not define structural or successional stages for this type, but just use a basic indication of grazing intensity:
40101--decent shape (no evidence of well-worn cattle trails)
40102--grazed by cattle (with bare soil, trails visible)
CONIFER FORESTS
Tree species composition is categorized as either mixed conifer, or as a tree species type if more than 80 percent of the canopy consisted of one (or an associated group) of conifer species (Douglas-fir, ponderosa pine, lodgepole pine, western larch, grand fir, Engelmann spruce, limber pine, a combination of western red cedar and western hemlock, western white pine, and a combination of subalpine fir, whitebark pine, and other subalpine tree species).
We define seven successional stages: 1) recent disturbance, 2) low shrub, 3) tall shrub, 4) pole/sapling, 5) young, 6) mature, and 7) old growth. For the first four stages we designated a structural type (following disturbance). By the time the stand could be defined as young forest, the original structure following disturbance is considered less important and is usually unknown, so it is no longer designated. We also have a common and diverse partially cut category (including thinning, single-tree selection, "sanitation" cuts, etc.), which, because of the complexity of the stands, we do not assign to a successional stage. Our classification of disturbed forest types is based on the existing plant species composition and stand structure (what the birds respond to) without regard to the process that actually caused the structure. Names associated with structural types (patch cut, shelterwood, seed tree, clearcut) are, therefore, merely descriptive of the process that probably (but not necessarily) gave rise to the current stand condition.
402xx - PONDEROSA PINE (>80%) - At least 80 % of the canopy cover is ponderosa pine.
403xx - DOUGLAS-FIR (>80%) - At least 80 % of the canopy cover is Douglas-fir.
404xx - WESTERN LARCH (>80%) - At least 80 % of the canopy cover is western larch.
405xx - MIXED CONIFER (no single species comprises >80% of the canopy cover) - No single conifer species makes up more than 80 % of the canopy cover in this forest type.
406xx - LODGEPOLE PINE (>80%) - At least 80 % of the canopy cover is lodgepole pine.
407xx - SPRUCE (>80%) - At least 80 percent of the canopy cover is Englemann spruce. Previously 40702 was a spruce/subalpine fir forest. Now the drier stands dominated by subalpine fir are classified with 408xx below, and only stands dominated by spruce are included here. These are usually more mesic and often at lower elevation.
408xx - SUBALPINE FOREST (mix of subalpine fir, lodgepole, spruce, or whitebark pine) Virtually all of the canopy cover is comprised of a combination of subalpine fir, lodgepole pine, spruce, whitebark pine, or subalpine larch (but NOT > 80 % lodgepole or > 80 % spruce). This forest type will most commonly be higher elevation and will usually have predominantly subalpine fir along with 1 or more of the other species. Until 2002, these would have been classified as mixed conifer (40502) so many points may be classified differently now.
409xx - GRAND FIR (>80%) - At least 80 % of the canopy cover is grand fir.
410xx - CEDAR-HEMLOCK (>80% cedar and/or western hemlock) - At least 80% of the canopy cover is comprised of a combination of cedar and hemlock (or either alone).
411xx - LIMBER PINE (>80%) - At least 80 % of the canopy cover is limber pine.
412xx - WHITE PINE (>80%) - At least 80 % of the canopy cover is western white pine.
413xx - MOUNTAIN HEMLOCK - At least 80 % of the canopy cover is mountain hemlock.
415xx - UPLAND DECIDUOUS (>80%) - At least 80 % of the canopy cover is comprised of aspen, birch, or sometimes cottonwood (but only in an upland setting). This code should be used for reasonably large patches (>20 m x 20 m) of aspen. A few scattered aspen are not enough to constitute a cover type (but note as a "feature").If there are many aspens interspersed with conifers, or multiple small patches, use 420xx.
420xx - MIXED CONIFER-DECIDUOUS MOSAIC (neither constitutes a stand by itself). This should be used for areas in which conifer and deciduous (usually aspen or birch) trees intermingle (>10-20% deciduous). Instead of distinct patches of each cover type, the 2 types of trees are scattered together individually or in small patches.
Successional stages within above forest types:
4xx01--old-growth - Usually a complex stand with trees of many age and size classes, large trees, snags, and usually downed woody debris (but usually not in ponderosa or larch old growth).
4xx02--mature - Often multi-storied with larger trees 20-40 cm dbh (depending on the species). The stand has self-thinned from the dense structure present in a young stand.
4xx03--young (may have little understory shrubs, canopy usually closed, most trees < 20 cm dbh, less heterogeneity than “mature”) Relatively dense and uniform stand of trees (10-20 cm dbh) that have regenerated following a clearcut or fire. Young stands generally (but not always) have a relatively closed canopy for any particular cover type (except ponderosa pine) and little understory. Trees are reached maybe only ½ of their full height for that species growing under similar conditions (elevation, aspect). A young forest is sometimes hard to distinguish. These stands have retained some features of the pole sapling successional stage, but have outgrown that category. Consider tree height and diameter, the structure of stand, and compare to a typical mature stand of the species in question. Do not pay attention simply to dbh – at higher elevations and exposed sites trees grow more slowly. Be aware that many mature lodgepole stands include small-dbh trees (and therefore would be classified as mature even with dbh < 20 cm).
With respect to the cover type classifications, stands with a history of timber harvesting often pose a problem because of the continuum of harvest types, particularly where harvesting has occurred more than once. For our purposes, base your classifications on the current stand structure (i.e., what the birds are responding to).
4xx04--thinned ( more open than mature, often multi-story) - A relatively small proportion of medium or large trees were selectively removed (the forest was thinned), as evidenced by the presence of stumps. The forest is otherwise mature, and may have a multi-aged, multi-storied structure, especially if cutting is not recent. More homogenous than group selection cut but usually more heterogeneous than shelterwood cut.
4xx34--overstory removal (most mature trees removed; small trees remain) - Most mature trees removed. Generally small trees up to 20+ cm remain, often clumped and interspersed with bare spots from logging activities, thus distinguishing it from a young forest following natural disturbance. A few larger trees may be left, especially if unmerchantable (deciduous, deformed, etc.), and the understory may be opened up to varying degrees.
4xx05-- patch cut; recent
4xx06-- patch cut; low shrub (< 2m) / seedling 2nd-growth stage
4xx07-- patch cut; tall shrub (>2m) / sapling-seedling 2nd-growth stage
4xx08-- patch cut; pole-sapling stage
This includes group-selection cuts or any other cut where groups of trees are removed here and there, leaving the appearance of an unnatural mosaic (significant patches of intact forest with significant areas of open-space between). This heterogeneous harvest practice is relatively new and may not be common on the landscape. Cuts should be at least 20 x20 m?
4xx09--shelterwood cut; recent
4xx10--shelterwood cut; low shrub (<2m) / seedling 2nd -growth stage
4xx11--shelterwood cut; tall shrub (>2m) / sapling-seedling 2nd-growth stage
4xx12--shelterwood cut; pole-sapling stage –
Large percentage of trees removed, but the mature, overstory trees that are left (sometimes to provide shade for seedlings) are uniformly spaced such that the stand appears orchard-like.
4xx13--seed tree cut; recent
4xx14--seed tree cut; low shrub (<2m)/seedling 2nd-growth stage
4xx15--seed tree cut; tall shrub (>2m)/sapling-seedling 2nd-growth stage
4xx16--seed tree cut; pole-sapling stage –
Most trees removed. A few widely spaced (usually mature) trees left to act as a seed source.
4xx17--clearcut; recent
4xx18--clearcut; low shrub (<2m)/seedling 2nd-growth stage
4xx19--clearcut; tall shrub (>2m)/sapling-seedling 2nd-growth stage
4xx20--clearcut; pole-sapling stage –
Nearly all trees removed, with only an occasional snag or spindly tree or a smattering of (generally small) trees left. This is very different from the numerous standing dead trees in the "natural" post-fire situation.
4xx21--post-fire (standing dead); recent (<5yr)
4xx22--post-fire (standing dead); low shrub (<2m)/seedling 2nd-growth stage
4xx23--post-fire (standing dead); tall shrub (>2m)/sapling-seedling 2nd-growth stage
4xx24--post-fire (standing dead); pole-sapling stage
This type is comprised of forest stands that have experienced a relatively intense stand-replacement crown fire, as evidenced by the presence of numerous, standing-dead snags. If virtually all snags have been removed by salvage-logging, classify as clearcut.
Bird’s eye view of various harvest techniques:
 |
 |
 |
 |
| Seed tree | Shelterwood | Group selection | Thinned |
RIPARIAN VEGETATION
50000 - LAKE, POND - Lake should be at least 20 m in diameter (Otherwise note as "feature").
50100 – MARSH - Marsh (50100) – has emergent vegetation (e.g., cattails, rushes, or sedges) associated with it. Standing water (possibly small pond) is usually present.
50200 - SEDGELAND, WET MEADOW, BOG - Wet meadow/sedgeland/bog (50200) – has grasses, some sedges and small forbs and standing water or soggy ground. Usually associated with conifer forest or above treeline.
503xx - WILLOW FLATS - Riparian areas dominated by willows, usually in flatter valleys.
50301--decent shape (lush understory, no visible browse line)
50302--grazed by cattle (degraded understory, sharp browse line)
504xx - COTTONWOOD BOTTOMLAND
50401--decent shape (lush understory, no visible browse line)
50402--grazed by cattle (degraded understory, sharp browse line)
505xx - STREAMSIDE RIPARIAN
50501--decent shape (lush understory, no visible browse line)
50502--grazed by cattle (degraded understory, sharp browse line)
This type includes drainage areas that have a well developed riparian vegetation component dominated by shrubs (for example, alder, birches, dogwood, hawthorn, etc.; NOT willows). Running water does not need to be present.
506xx - HARDWOOD DRAW - most common in eastern Montana and the Dakotas.
50601--decent shape (lush understory, no visible browse line)
50602--grazed by cattle (degraded understory, sharp browse line)
OTHER
60xxx – TUNDRA - tundra is high elevation but does not necessarily have to be above tree line.
60100--shrub tundra
60200--herbaceous tundra
70000 - BARREN LAND/TALUS - rock or bare soil, few or no trees or shrubs.
Previously, this was only barren land at high elevation. Now this can be used for any elevation. In 2000, talus slope (85000) was added as a separate code, but this is now included here.
80000 - CLIFFS/ROCK WALLS - more extensive than a rock outcrop.
Mosaic – Is the habitat part of a larger mosaic of 2 different cover types? (Hint - refer to aerial photo for an idea of the larger context).
The mosaic option is not meant to be used very often. It does NOT refer to heterogeneity or other types of "messes." It also does not mean a "mosaic" of many different cover types. It refers to a fairly regular pattern of two cover types interspersed together, at a fine enough grain that it seems inappropriate to classify it as two separate things. It is meant to capture combinations that occur fairly often in some areas (such as ponderosa pine and grassland, or Douglas-fir and sagebrush) and that may be important to some bird species.
Mosaic clarification (example)
 |
 |
 |
| forest mosaic | open p-pine forest | forest with grassland edge |
The remaining vegetation measurements are taken away from the road or trail. In order to characterize the vegetation structure and species composition of the habitat you designated in COVTYPE, you will need to be well within this habitat and away from the influence of the road: Walk perpendicular to the road (turning right from the direction facing the start point), until you reach a point that is 30 m inside the boundary of typical habitat, i.e., uninfluenced by the road. In those instances where the edge of some other cover type, i.e., Edge1 or Edge2, occurs within 30 m of this new location, you should shift the center of your vegetation plot, or try the other side of the road, so that it lies completely within the recorded (COVTYPE). You may be able to anticipate the need for such a shift from the road. If you cannot fit a 30-m-diameter circle within the primary cover type, then measure the vegetation within that portion of the circle that is within the cover type (this will be especially necessary for narrow riparian strips). Vegetation cover estimates should take into account all heterogeneity within the circle (or portion of the circle) of COVTYPE habitat.
The next series of variables (Table 4) are estimated from within an area defined by a 30-m-radius circle centered on the off-road vegetation point. You first need to determine if you have a "canopy" layer. "Canopy" is defined as the vegetation layer containing medium and large trees (>20 cm dbh), which constitutes an overstory, including 10-20 cm "poles," as long as they are clearly within the canopy or high subcanopy (not understory) layers. Do not include understory trees (saplings and seedlings). If most trees in the upper layer are < 10 cm dbh, then there is no canopy layer (canopy cover = 0), and the coverage of these trees should be included in sapling cover.
In addition to the visual estimates of percent cover of each vegetation height class, record names of the dominant species (those that make up > 85-90% of the total coverage). For the canopy layer only, estimate the percent of the total canopy coverage that is represented by each tree species listed. The general kind of shrub and bush (genus) is enough information for those layers; grasses and forbs need no identification. If an important plant species is unknown, take a small sample to identify later.
If there is a canopy layer, record tree height as the typical height of the overstory, excluding emergent trees. If only a few large trees remain in a harvest unit, e.g., scattered "seed" trees, estimate the typical height of those trees only. If only seedlings, saplings, or shrubs exist within 30 m, record height of highest layer, and percent canopy coverage as "0". Use a clinometer to measure tree height. Once confident of height classes, height may be estimated from one point to another. However, it is extremely important to calibrate your estimates periodically with the use of a clinometer.
Table 5. Vegetation variables measured within a 30-m-radius circle surrounding the count point.
VARIABLE |
EXPLANATION |
CANOPY |
Visually estimate the percent canopy cover (0 to 100%) within 30 m. The "canopy" is comprised of a layer of trees that is usually > 20 cm dbh, but includes trees > 10 cm that are in the canopy or high subcanopy layers. |
SAPLING |
Visually estimate the percent sapling coverage (0 to 100%) within 30 m. Saplings are defined as trees between 5-10 cm dbh, or a slightly larger diameter tree that is clearly in the sapling (not canopy) layer. |
SEEDLING |
Visually estimate the percent seedling coverage (0 to 100%) within 30 m. Seedlings are defined as trees < 5 cm dbh |
SHRUB |
Visually estimate the percent shrub coverage (0 to 100%) within 30 m. Shrubs are defined as multi-stemmed woody plants greater than 1 m tall. |
BUSH |
Visually estimate the percent short-shrub coverage (0 to 100%) within 30 m. Short shrubs (bushes) are defined as those multi-stemmed, woody plants less than 1 m tall. (Estimate by clump, if predominantly greater than or less than 1 m.) |
GRASS |
Visually estimate the percent grass/forb cover (0 to 100%) within 30 m. |
HEIGHT |
Estimate the typical height of the highest layer of trees within 30 m. If you have designated a "canopy" layer (above), the height should be the typical height of that layer, excluding occasional overdominant (emergent) trees. If there is no canopy layer, record the typical height of the highest significant layer of trees or shrubs. Note that scattered seed trees do comprise a canopy layer. Provide an estimate to the nearest m. |
SNAGS |
Estimate the abundance of snags (> 20 cm dbh) within 30 m. Stumps shorter than breast height are placed in the next category. Code with a "0" for none, "1" for 1-3, "2" for >3 snags. |
STUMPS |
Estimate the abundance of stumps (.5-1.5 m high and >20 cm diameter) within 30 m. Code with a "0" for none, "1" for 1-5, "2" for >5 stumps. |
DOWNED |
Estimate the abundance of logs and other woody debris ( >20 cm diameter) within 30 m. Code with a "0" for none, "1" for 1-5, "2" for >5 items. |
Three vegetation variables are recorded within 15 m of the vegetation point (Table 6).They are intended to sample the density and size distribution of trees in the canopy layer.You should only include those trees that are within the layer that you referred to as canopy in your cover estimates (above). Pole-sized trees (10-20 cm) may or may not be in that layer. Use a dbh (diameter at breast height) tape to classify trees into one of three dbh classes for stem counts. Alternatively, if you know your finger spread for each size class, you can quickly estimate dbh of trees this way (practice on known-size trees until you can correct for parallax). However, it is extremely important to calibrate your estimates periodically with the use of a dbh tape.
Table 6. Vegetation variables measured within 15 m of the count point.
VARIABLE |
EXPLANATION |
LGTREE |
Count the number of large-dbh (>40 cm) trees within 15 m of count point |
MEDTREE |
Count the number of medium-dbh (20-40 cm) trees within 15 m of point |
POLETREE |
Count the number of pole-dbh (10-20 cm) trees within 15 m of point that are in the canopy or subcanopy layers |
Before leaving the point, make sure that (1) all boxes are filled in; (2) values are internally consistent (e.g. cover type and tree species composition agree); and (3) the point is placed accurately on the map and aerial photo.
Data should be transposed from field record sheets to the computer on a daily basis, if possible. That way, the data can be viewed while one's memory of the day's events is fresh. The bird data are entered into a separate file from the vegetation data to save computer space (see formats in Appendix IV). The files are then combined later, so it is important that the key variables that are used to link the two files (observer, date, stop, transect) are identical.Even when bird and vegetation sampling for a transect are done on different days, or by different observers, enter the same bird count date and observer in both files. You should enter the vegetation data first, and then make a new file with just the key variables that are repeated in the bird data file. Use this new file to enter the bird data.
If you enter something different from the data book for any reason: first, be very sure it is correct, and second, explain the discrepancy in the book (or change it). Otherwise we may assume it is a data entry error and change it in the computer.
Data are entered in a standard spreadsheet format. All national forests have standard spreadsheets available on PCs, Excel being the preferred format.
It is extremely important that all data files forwarded to the data management center (currently the University of Montana) are in the exact format given in Appendix IV. The order and names of variables, variable types (alpha or numeric), column widths, and all codes must be exact. ENTER ALL ALPHANUMERIC DATA IN UPPER CASE - ALL CAPS!
Hutto, R. L. 1998. Using landbirds as an indicator species group. Pp. 75-92 in Marzluff, J. M., and R. Sallabanks (eds.), Avian conservation: Research and Management. Island Press, Covelo, CA.
Hutto, R. L., S. M. Pletschet, and P. Hendricks. 1986. A fixed-radius point count method for nonbreeding and breeding season use. Auk 103:593-602.
Hutto, R. L., S. J. Hejl, J. F. Kelley, and S. M. Pletschet. 1995. A comparison of bird detection rates derived from on-road versus off-road point counts in northern Montana. Pp. 103-110 in C. J. Ralph, J. R. Sauer, and S. Droege (tech. eds.) Monitoring bird populations by point counts. USDA For. Serv. Gen. Tech. Rep. PSW-GTR-149, Albany, CA
Hutto, R. L., and J. S. Young. 1999. Habitat relationships of landbirds in the USFS Northern Region. USDA Forest Service Gen. Tech. Rep. RMRS-32.
Ralph, C. J., J. R. Sauer, and S. Droege, technical editors. 1995. Monitoring bird populations by point counts. Gen. Tech. Rep. PSW-GTR-149. Albany, CA, 187 pp.
Young, J. S., and R. L. Hutto. 1998. Effects of partial-cut timber harvesting on landbirds in the northern Rocky Mountains. Report, in preparation for publication.
Young, J. S., and R. L. Hutto. 1999. Habitat and landscape factors affecting cowbird distribution in the Northern Rockies. Studies in Avian Biology 18: 41-51.
Young, J. S., and Hutto, R. L. 2002. Use of regional-scale exploratory studies to determine bird-habitat relationships. Pages 107-119 in J. M. Scott, P. J. Heglund, M. Morrison, M. Raphael, J. Haufler, and B. Wall, editors. Predicting Species Occurrences: Issues of Scale and Accuracy. Island Press, Covello, CA.
APPENDIX I. Mapping materials and field equipment.
____________________________________________________________________________________________
MATERIALS FOR MAPPING POINT AND TRANSECT LOCATIONS:
Forest travel map- One copy for each crew person of the most current forest road map available.
USGS original 7.5-minute topographic quads, blue-line quads, or ortho-photo maps which cover the forest - Original 7.5-minute topos are most easily photocopied for field use. Blue-line quads and ortho-photos may be necessary for locating recent roads and trails.
Color photocopies of recent color aerial photos - one aerial for each mapped transect.
Transect Location Forms (see Appendix II)
Random number table - available in most statistics books.
Map wheel
Ruler
Pencil
Highlighter
Fine-line permanent black ink pen
FIELD EQUIPMENT:
clipboard and plastic page cover
pencils; good erasers
daypack or cruiser vest
2' wooden or rebar stakes
1.25" round aluminum tree tags (blank) (10 per transect)
numbered die set (for marking tags 1-10)
aluminum tree nails for tags
hammer of appropriate weight for stakes
flagging - 1 roll/observer sufficient
azimuth compass
thermometer - Fahrenheit - with protective case
digital watch with timer
metric dbh tape (2-meter tape sufficient)
metric 30-meter measuring tape
camping equipment
binoculars
field guides
methods protocols and data forms
APPENDIX II. Transect location form
____________________________________________________________________________________________
TRANSECT LOCATION FORM
FOREST
TOPO QUAD NAME YEAR OF MAP MAP NUMBER
____________________________
TRANSECT NAME TRANSECT NUMBER
/ /
YEAR ESTABLISHED MO /DA /YR FIRST RUN
OBSERVER IN FIRST YEAR
START POINT:
/ /
TOWNSHIP RANGE SECTION
__ __ __ __ . _ __ __ __ __ . __
LATITUDE LONGITUDE
DIRECTIONS TO START POINT FROM NEAREST MAIN ROAD:
4WD ONLY 2WD OKAY HIGH CLEARANCE ONLY
ROAD CONDITION/ACCESS UNCERTAIN
HIKE IN TO START OPEN ROAD CLOSED ROAD
(Point location descriptions on back)
APPENDIX III. REVISIONS FORM
____________________________________________________________________________________________
FOREST DISTRICT
QUAD DATE OF CHANGE
OBSERVER
TRANSECT TRANSECT NO.
NAME
TYPE OF CHANGE/ADDITION MADE TO:
TRANSECT LOCATION FORM:
TOPOGRAPHIC MAP:
APPENDIX IV. Data file structure for bird and vegetation data. Variable names, types, and column widths for the bird and vegetation files must be submitted EXACTLY as described for all files sent to the data center: NO NAME CHANGES, ADDITIONS OR DELETIONS, PLEASE. ____________________________________________________________________________________________
BIRD FILE:
Variable Description Type Width
-------- -------------------------------------------- ------ -----
*FOREST USFS FOREST NUMBER NUMERIC 2
*TRANSECT UNIQUE TRANSECT NUMBER (LATILONG OF 1ST STOP) NUMERIC 9
*OBSERVER BIRD OBSERVER'S 3 INITIALS ALPHA 3
*DATE YR, MO, and DA OF BIRD COUNT (NO HYPHENS!) NUMERIC 6
*STOP STOP NUMBER ON TRANSECT NUMERIC 2
TIME TIME-OF-DAY THAT BIRD COUNT WAS BEGUN NUMERIC 4
WIND BEAUFORT WIND CONDITION CODE NUMERIC 1
SKY SKY CONDITION CODE NUMERIC 1
TEMP AIR TEMPERATURE NUMERIC 2
STREAM STREAM NOISE CODE NUMERIC 1
SPECIES MNEMONIC 4-LETTER BIRD SPECIES CODE ALPHA 4
ABUN NUMBER OF INDIVIDUAL BIRDS DETECTED NUMERIC 3
DISTANCE DISTANCE OF BIRD(S) FROM OBSERVER NUMERIC 3
CUES_A SOUND CUE USED TO IDENTIFY BIRD (IF ANY) ALPHA 1
CUES_V VISUAL CUE USED TO IDENTIFY BIRD (IF ANY) ALPHA 1
LOCATION WHETHER BIRD WAS WITHIN MAIN COVER TYPE NUMERIC 1
_______________________________________________________________________