J.-F. Therrien, L. J. Goodrich, D. R. Barber, and K. L. Bildstein. 2012. A long-term database on raptor migration at Hawk Mountain Sanctuary, northeastern United States. Ecology 93:1979. http://dx.doi.org/10.1890/12-0353.1

METADATA

CLASS I. DATA SET DESCRIPTORS

A. Data set identity: Long-term counts of 18 species of diurnal birds of prey migrating past Hawk Mountain Sanctuary in autumn, in the Central Appalachian Mountains of eastern Pennsylvania, northeastern United States.

B. Data set identification code: HMS_Migration_Data

C. Data set description: Ongoing autumn migration counts of 18 species of diurnal birds of prey. In most years, the counts tally between 15,000 and 25,000 birds of prey between 15 August and 15 December, annually. The database is updated annually.

Principal Investigator: Jean-François Therrien, Acopian Center for Conservation Learning, Hawk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961, USA. Queries regarding the data set can be directed to: [email protected]

Abstract: In most northern temperate regions diurnal birds of prey, or raptors, migrate seasonally between their breeding and wintering grounds. Although their populations can be logistically difficult to survey and monitor because they are largely secretive and wide-ranging, most raptors are obligate or facultative soaring migrants that congregate along major thermal and orographic updraft corridors during their seasonal movements. Hawk Mountain Sanctuary (41° N, 75° W), which straddles the Kittatinny Ridge, the southernmost ridge in the Appalachian Mountains in eastern Pennsylvania, witnesses large numbers of migrating raptors during autumn migration. The Sanctuary’s long-term migration-count database is the oldest detailed archive on the timing and magnitude of migratory raptors in the world. Records comprise daily (1934–1965) or hourly (1966–present) counts of 18 North American species of raptors migrating past the Sanctuary, as well as detailed weather data recorded from the lookout during autumn migration (15 August to 15 December). The long-term data set has been valuable in understanding raptor migration and population trends. Because top predators such as raptors are sensitive bio-indicators of ecosystem changes, variations in the numbers of individual species also may reflect changes in the health of the environment. Thus, monitoring population trends of raptors can provide crucial information on environmental changes and threats facing wildlife species. Indeed, the Sanctuary’s database has been key in identifying the now well-known example of organochlorine contamination of top predators and has helped the conservation, management, and reintroduction of many species such as Osprey (Pandion haliaetus), Bald Eagle (Haliaeetus leucocephalus), and Peregrine Falcon (Falco peregrinus). This database is of prime significance to conservation ecology and long-term fluctuations in numbers of migratory birds.

D. Key words: bird counts; bird migration; Hawk Mountain Sanctuary; long-term studies; Pennsylvania (USA); population census; raptor.

CLASS II. RESEARCH ORIGIN DESCRIPTORS

A. Overall project description

Identity: Year, month, day, time of day, daily (1934–1965) or hourly (1966–present) counts of individual raptors (18 species) migrating past the Sanctuary’s North Lookout, weather data (visibility, cloud cover, air temperature, wind speed, wind direction), number of observers conducting the count, and exact number of minutes observed during specified hour.

Originator: Maurice Broun (deceased), 1934.

Period of Study: 1934–present, except for 1943–1945.

Objectives: Monitoring post-breeding population trends of 18 North American diurnal raptor species migrating through northeastern United States.

Abstract: same as above

Sources of funding: Hawk Mountain Sanctuary Association and its supporters.

B. Specific subproject description

Site description: Hawk Mountain Sanctuary straddles the Kittatinny Ridge, in the Central Appalachian Mountains of eastern Pennsylvania. The North Lookout (40°58'N, 74°59'W) is 464 m above sea level and provides unobstructed view 200° facing northeast.

Site type: Hawk Mountain Sanctuary is predominately covered with mature deciduous and coniferous forests. Surrounding lands are a mixture of forested, agricultural, and small villages. The ridge has a northeast-southwest orientation that acts as a leading line (sensu Bildstein 2006) for migrating raptors (Bildstein 1999). Migrating raptors mostly follow this direction as they pass the North Lookout.

Geography: The Sanctuary’s 10 km² is bordered by the Little Schuylkill River to the west.

Habitat: Most of the Sanctuary’s habitat is second-growth oak (Quercus sp.), maple (Acer sp.), eastern hemlock (Tsuga canadensis), and pine (Pinus spp.) forest.

Geology: Hawk Mountain Sanctuary and the surrounding Appalachian Mountains date to the Late Cambrian (500 million years ago). Details on geology are in Brett (1986).

Watersheds/hydrology: The Sanctuary is part of the Schuylkill River watershed.

Site history: Given the large numbers of migrating raptors following the Kittatinny Ridge each autumn, the uppermost outcrops of the ridge were well-known at the beginning of the 20th century by local hunters as a great place to shoot raptors, which were considered “vermin” at that time. Hawk Mountain Sanctuary was founded in 1934 by conservationist Rosalie Edge as the world’s first refuge for birds of prey. Edge established the Sanctuary to stop the shooting of migrating hawks there (Broun 1949, Bildstein 2006). In the autumn of 1934, counts of migrating raptors were established as a way to document the number of raptors benefiting from the Sanctuary’s protection.

Climate: Hawk Mountain Sanctuary has a continental temperate climate with monthly average temperatures varying from -3.4°C during the coldest month (January) to 21.4°C during the warmest month (July). Mean annual rainfall equals 958 mm.

Experimental design: Data are the result of an ongoing 75-year observational field study.

Design characteristics: N/A

Sampling methods: One or more experienced (in raptor flight identification) observers scan the sky from north to south horizon with 8X to 10X binoculars and unaided eye to spot migrating raptors. Observers use a 20 by 60X telescope to identify distant birds when needed, but the counts are restricted to birds spotted with binoculars. Numbers of birds were recorded daily from 1934 to 1965. Since 1966, numbers of birds have been recorded hourly. Individual raptors are counted if they are clearly moving south or southwest past the North Lookout (Bednarz et al. 1990, Allen et al. 1996). Counts are conducted (weather permitting) each day from 15 August to 15 December, generally between 0800 to 1700 EST. A mix of highly trained observers, most of them with over 10 years of experience in raptor flight identification, rotating through the days insures consistency of data collection. Excluding the 1943–1945 period when counts were suspended because of World War II, specific hours or days without data are almost exclusively the result of adverse climatic conditions. Indeed, thick fog, heavy rain and/or snowfall can all reduce visibility to null, thus preventing the counts.

Taxonomy and systematics: Migrating raptors are identified as precisely as possible according to the 28 classes listed below.

Permit history: N/A

Legal/organizational requirements: N/A

Project personnel: Currently, Laurie J. Goodrich, David R. Barber, Jeremy Scheivert, Jean-François Therrien, Keith L. Bildstein, and dozens of trained volunteers.

CLASS III. DATA SET STATUS AND ACCESSIBILITY

A. Status

Latest update: The data span the period of 1934–2010 (except for 1943–1945). The database is updated annually.

Latest Archive date: July 2011.

Metadata status: The metadata are complete and up to date.

Data verification: Field data were entered into an ASCII (.txt) file. These data entries were then checked for typographical and other data-entry errors.

B. Accessibility

Storage location and medium: The original ASCII (.txt) file is stored on computers at the Acopian Center for Conservation Learning, Hawk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961, USA.

Contact person: Jean-François Therrien, Acopian Center for Conservation Learning, Hawk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961, USA. Queries regarding the data set can be directed to: [email protected]

Copyright restrictions: None.

Proprietary restrictions: None.

Costs: None.

CLASS IV. DATA STRUCTURAL DESCRIPTORS

A. Data Set File

Identity: HMS_Migration_Data.txt

Size: This file contains the complete migration records for the period 1934–2010 (42,000 entries). Each row consists of 44 cells (described below).

Format and storage mode:Data are stored in an ASCII (.txt) file, comma delimited. No compression was used.

Header information: See variable names in Section B.

Alphanumeric attributes: Mixed.

Special characters/fields: Missing data denoted as ".".

B. Variable information

Data in the file HMS_Migration_Data.txt are organized laterally in rows. From 1934 to 1965, each row of the data represents one day of observation. Since 1966, each row of the data represents one hour of observation. Variables are in columns. Variables in the ASCII (.txt) file (= columns in table format) are:

ID: A single number is given for each individual row (starting from 1 to 42,000)

Year: numbers indicate the year, from 1934 to 2010.

Month: numbers indicate the month, i.e., 9 = September, 10 = October, etc...

Day: numbers indicate the day e.g., 6 = 6th of the respective month.

Time: numbers indicate time of day (Eastern Standard Time formatted in 24-hr notation as hhmm).

Max visibility: estimated maximum distance (km) seen clearly from the lookout during observation period determined by interpolation between landmarks of known distances ranging from 0 to 77 km.

Cloud cover: estimated proportional amount of cloud covering the sky (measured since 2006 only).

Air temperature: air temperature (°C).

Sky code: predominant sky conditions during observation period according to the following codes:

1. Clear, 0–15% cloud cover
2. Partly cloudy, 16–50% cloud cover
3. Mostly cloudy, 51–75% cloud cover
4. Overcast, 76–100% cloud cover
5. Wind-driven sand, dust, snow
6. Fog or haze
7. Drizzle
8. Rain
9. Snow
10. Thunderstorm, with or without precipitation

Wind speed code: predominant wind speed during observation period according to the Beaufort scale codes:

1. Less than 1km/h, smoke rises vertically
2. 1–5 km/h, smoke drift shows wind direction
3. 6–11 km/h, leaves rustle, wind is felt on face
4. 12–19 km/h, leaves, small twigs in constant motion
5. 20–28 km/h, raises dust, leaves loose paper, small branches in motion
6. 29–38 km/h, small trees in leaf sway
7. 39–49 km/h, longer branches in motion
8. 50–61 km/h, whole trees in motion
9. 62–74 km/h, twigs, small branches broken off trees
10. Greater than 75 km/h

Wind from: direction the wind is coming from during observation period using one (N, S, E, W), two (NE, SW…) or three-letter (NNW…) abbreviation. We use the three-letter code “VAR” for variable wind.
Altitude flight code: estimated predominant height above local topography at which the birds are passing during observation period according to the following codes:

1. Below eye level
2. Eye level up to about 30 meters overhead
3. Birds easily seen with unaided eye
4. At limit of unaided vision
5. Beyond limit of unaided vision, but visible with binoculars (10X)
6. At limit of binoculars (10X)
7. Beyond limit of binoculars (10X)
8. No predominant height

Flight direction: direction in which the birds are flying as they pass the lookout during observation period using one (N, S, E, W), two (NE, SW…) or three-letter (NNW…) abbreviation. We use the three-letter code “VAR” for variable direction.

Observers: number of people actively and reliably contributing to the count during observation period.

Minutes: exact number of minutes of observation during observation period.
BLVU: number of individual black vultures (Coragyps atratus) observed during observation period.
TUVU: number of individual turkey vultures (Cathartes aura) observed during observation period.
UNVU: number of individual unidentified vultures (turkey or black vulture) observed during observation period.
OSPR: number of individual ospreys (Pandion haliaetus) observed during observation period.
BAEAU: number of individual bald eagles (Haliaeetus leucocephalus) of undetermined age observed during observation period.
BAEAI: number of individual immature bald eagles observed during observation period.
BAEAA: number of individual adult bald eagles observed during observation period.
NOHA: number of individual northern harriers (Circus cyaneus) observed during observation period.
SSHA: number of individual sharp-shinned hawks (Accipiter striatus) observed during observation period.
COHA: number of individual Cooper’s hawks (A. cooperii) observed during observation period.
NOGO: number of individual northern goshawks (A. gentilis) observed during observation period.
UACC: number of individual unidentified Accipiter (sharp-shinned hawk, Cooper’s hawk or northern goshawk) observed during observation period.
RSHA: number of individual red-shouldered hawks (Buteo lineatus) observed during observation period.
BWHA: number of individual broad-winged hawks (B. platypterus) observed during observation period.
RTHA: number of individual red-tailed hawks (B. jamaicensis) observed during observation period.
RLHA: number of individual rough-legged hawks (B. lagopus) observed during observation period.
SWHA: number of individual Swainson’s hawks (B. swainsoni) observed during observation period
UBUT: number of individual unidentified Buteo (red-shouldered hawk, broad-winged hawk, red-tailed hawk, rough-legged hawk or Swainson’s hawk) observed during observation period.
GOEAU: number of individual golden eagles (Aquila chrysaetos) of undetermined age observed during observation period.
GOEAI: number of individual immature golden eagles observed during observation period.
GOEAA: number of individual adult golden eagles observed during observation period.
UEAG: number of individual unidentified eagles (bald eagle or golden eagle) observed during observation period.
AMKE: number of individual American kestrels (Falco sparverius) observed during observation period.
MERL: number of individual merlins (F. columbarius) observed during observation period.
PEFA: number of individual peregrine falcons (F. peregrinus) observed during observation period.
GYRF: number of individual gyrfalcons (F. rusticolus) observed during observation period.
UFAL: number of individual unidentified falcons (American kestrel, merlin, peregrine falcon or gyrfalcon) observed during observation period.
UNRAP: number of individual unidentified raptors (any species from the above list) observed during observation period.
TOTRAP: sum of individual raptors (all species combined) observed during observation period.

CLASS V. SUPPLEMENTAL DESCRIPTORS

A. Data acquisition

Data forms: Field notebooks (1934–1965) and paper field sheets (1966–present).

Location of completed data forms: The original field notes collected between 1934 and 1965 are stored at Muhlenberg College, Allentown, PA 18104, USA. Copies of the 1934–1965 field notes and original field sheets from 1966 to present are stored at the Acopian Center for Conservation Learning, Hawk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961, USA.

Data entry/verification procedures: The original data were recorded by hand into field notebooks (1934–1965) and field sheets (1966-present). The data were entered into an electronic ASCII (.txt) file. All data entries were checked against field sheets for each observation period.

B. Quality assurance/quality control procedures:

Counts for the period 1934–2002 (before the extensive use of computers) were carefully checked and confirmed when first entered in the database. Since 2003, all data are entered daily at the end of each day by the official counter of the day and later inspected by the database manager. For each daily count entered in the database, summations of all species are back checked with field observations. We search the database annually for outliers and potential errors by filtering data in each variable. We then compute daily, monthly, and yearly averages and summations and look at minimum and maximum values to make comparisons with historical records. If any irregularities are found, we search original field sheets or notebooks to revise or confirm data. Moreover, all apparently extreme weather values are verified for accuracy using weather archive data from Pennsylvania State University (http://climate.met.psu.edu) and confirmed by a weather station located in Reading, Pennsylvania, less than 30 miles from the Sanctuary.

C. Related material: N/A

D. Computer programs and data processing algorithms: N/A

E. Archiving: N/A

F. Publications using the data set:

A near-complete list of publications using the data set can be found at: http://www.hawkmountain.org/science/scientific-publications/scientific-publications/page.aspx?id=323

G. Publications using the same sites:

See above.

H. History of data set usage

The long-term data set of autumn migration recorded at Hawk Mountain Sanctuary has been valuable in understanding raptor migration patterns and behavior (Allen et al. 1996, Maransky et al. 1997) and population trends (Nagy 1977, Bednarz et al. 1990, Viverette et al. 1996, Farmer et al. 2007, 2008, Farmer and Smith 2009). The data recorded over the years have documented the now well-known example of organochlorine contamination of birds of prey (Carson 1962), and have helped the conservation, management and reintroduction of many species such as Osprey, Bald Eagle, and Peregrine Falcon (Hickey 1969, Bildstein and Compton 2000, Bildstein 2006, Bildstein et al. 2008).

Data request history: N/A

Data set update history: N/A

Review history: N/A

Questions and comments from secondary users: N/A

ACKNOWLEDGMENTS

We thank the many trained volunteers and staff observers that have conducted migration counts over the years and the late Maurice Broun who recorded most counts between 1934 and 1965. This is Hawk Mountain Sanctuary contribution to conservation science number xxx.

LITERATURE CITED