Red-shouldered Hawk and Spring Woodpecker Survey

2003 Final Report

Produced For Ontario Ministry of Natural Resources – Wildlife Assessment Program

January 2004

Debbie S. Badzinski

Bird Studies Canada / Études D’Oiseaux Canada P.O. Box/B.P. 160, 115 Front St., Port Rowan, ON Canada N0E 1M0 www.bsc-eoc.org Page 1 of 25

EXECUTIVE SUMMARY

The Red-shouldered Hawk (Buteo lineatus) is a rare to locally uncommon breeding bird in Ontario (Austen and Cadman 1994, Austen et al. 1994, James 1991, Sutherland 1994), and is classified as a species of Special Concern by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) and Vulnerable in Ontario by the Ontario Ministry of Natural Resources (OMNR). The Red-shouldered Hawk survey began in 1990, with the primary goal of assessing whether forest management practices were negatively affecting Red-shouldered Hawks in Ontario. Surveyors also gather data on other raptor species and five species of woodpeckers.

In 2003, 53 volunteers completed 55 Red-shouldered Hawk and Spring Woodpecker survey routes throughout central Ontario. Surveyors recorded 167 Red-shouldered Hawks on 67% of the routes, for a mean of 3.04 birds/route. This is the second lowest number of Red-shouldered Hawks recorded since the survey began. Turkey Vulture was the second most common raptor species recorded by volunteers (mean: 1.38 birds/route), followed by Broad-winged Hawk (mean: 1.07 birds/route) and Red-tailed Hawk (mean: 0.22 birds/route). Average numbers of each raptor species have shown little change between 1991 and 2003.

Despite the observed decline in number of Red-shouldered Hawks detected during the 2003 survey, numbers have been stable with only minor annual fluctuations from 1991-2003. This suggests that although the population remains small and is considerably lower than it was historically, there is sufficient habitat in central Ontario to maintain the present Red-shouldered Hawk population. However, the continued expansion of urban areas and ever-increasing cottage development in prime Red-shouldered Hawk breeding habitat are ongoing threats to the Ontario population.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 2 of 25

TABLE OF CONTENTS

EXECUTIVE SUMMARY...... 1

INTRODUCTION ...... 3

METHODS...... 4

Broadcast equipment...... 6

Route selection ...... 6

DATABASE STRUCTURE ...... 7

DATA ANALYSIS...... 8

RESULTS...... 9

Volunteer participation ...... 9

Survey conditions ...... 9

Survey results ...... 9

Red-shouldered Hawks ...... 9

Other raptors...... 13

Woodpeckers...... 13

Woodpecker trends from Breeding Bird Survey ...... 19

DISCUSSION...... 19

Red-shouldered Hawk population changes...... 19

Woodpecker population changes...... 20

ACKNOWLEDGMENTS...... 20

LITERATURE CITED ...... 21

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 3 of 25

INTRODUCTION

The Red-shouldered Hawk (Buteo lineatus) was once the most common hawk in southern Ontario, but it has experienced a dramatic, continent-wide decline in the last century (Crocoll 1994). The Red-shouldered Hawk is now a rare to locally uncommon breeding bird in Ontario (Austen and Cadman 1994, Austen et al. 1994, James 1991, Sutherland 1994), and is classified as a species of Special Concern by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) and Vulnerable in Ontario by the Ontario Ministry of Natural Resources (OMNR). These designations are largely based on the decline in the southern Ontario breeding population that has occurred in the last century (Risley 1982, Austen and Cadman 1994). Ninety percent of Ontario breeding records for Red-shouldered Hawks are from the Great Lakes Forest Region, and a few additional breeding sites have been identified in the Carolinian and Southern Boreal forests. Ontario constitutes approximately 5% of the Red-shouldered Hawk’s total North American breeding range (Austen and Cadman 1994).

Red-shouldered Hawks occur in a broad array of North American forest types, but they prefer large tracts of mature to old growth, mixed forests, especially bottomland hardwood riparian areas and flooded deciduous swamps (Crocoll 1994). This species also requires closed canopy forests for successful reproduction, and is sensitive to changes in forest structure following timber harvesting (Crocoll 1994). Conversely, reduced canopy closure as a result of logging practices may benefit Red-tailed Hawks (Buteo jamaicensis), which may then displace or out- compete Red-shouldered Hawks (Bryant 1986). The dependence of Red-shouldered Hawks on large, mature forests raised the concern that timber management practices may be negatively affecting the hawks.

As a result, the Red-shouldered Hawk and Spring Woodpecker Survey was initiated in 1990. The primary objectives of the survey are to provide statistically reliable population trend data for Red-shouldered Hawks, Pileated Woodpeckers, and Yellow-bellied Sapsuckers and determine whether forest management practices are negatively affecting these species. The study area is restricted to deciduous or mixed forest habitat in central Ontario, the core breeding range of the Red-shouldered Hawk.

Woodpeckers in general, and more specifically the Pileated Woodpecker and Yellow-bellied Sapsucker, may also be influenced by forest management practices in Ontario, because these species rely on large, relatively mature forests and depend on snags for breeding. Pileated Woodpeckers and Yellow-bellied Sapsuckers are easier to detect in late April and early May (the current survey window for Red-shouldered Hawks) than in June when other breeding bird surveys are conducted (Peterjohn et al. 1994). Furthermore, because of their secretive habits during June, these species are not adequately monitored by other bird surveys, such as the Breeding Bird Survey (McLaren 1993). For this reason, participants in the Red-shouldered Hawk survey also collect data on woodpeckers.

The Red-shouldered Hawk and Spring Woodpecker Survey is run by Bird Studies Canada, as

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 4 of 25 part of the Ontario Birds At Risk (OBAR) program, in cooperation with the Wildlife Assessment Program of the Ontario Ministry of Natural Resources. This report summarizes results of the 2003 survey. Annual indices were calculated for numbers of Red-shouldered Hawks from 1991– 2003, and for the five most frequently encountered woodpecker species for the period 1992– 2003 (1996-2003 for non-target species).

METHODS

The Red-shouldered Hawk and Spring Woodpecker Survey is a roadside survey that uses playback calls to elicit responses from Red-shouldered Hawks. Observations of other raptor and woodpecker species are recorded by passive listening or visual observations. The survey methods were proposed by Szuba (1990), and were further tested and modified by Callaghan (1990) and Austen (1991).

Survey routes are located along secondary roads passing through as much deciduous woodland as possible. Each survey route is 19 km long and consists of 20 stations spaced 1 km apart. All routes are surveyed once annually within the time period of 17 April- 7 May. This survey window was selected to correspond with the pre-incubation period for Red-shouldered Hawks in Ontario, during which the highest response rate to playback is expected.

Prior to the survey, each surveyor is provided with a participant's kit containing: a 1:50,000 scale topographic map of their survey route; stop descriptions of their survey route; a participant's guide with survey instructions; a set of data forms (survey form, bird data form, stop description form); a broadcast tape or CD with Red-shouldered Hawk calls; a training tape or CD with examples of hawk calls and woodpecker calls and drumming; a dashboard sign; and a tax relief form. When the survey first began, all surveyors were provided with a broadcast unit with a 12- V car adapter cord. However, in recent years, the cost of providing broadcast units has become prohibitive and as a result, many surveyors use their own equipment.

Each year, new surveyors are asked to scout their route prior to their survey to establish survey stations and determine proper spacing based upon their vehicle’s odometer. If a potential survey station is located in an unsuitable location (e.g., no forest within 500 m, adjacent to a house or urban area), surveyors are instructed to move their stop an additional 500 m and assess the new location. This process is repeated until a suitable survey station is found. All survey stations must be at least 1 km apart and all routes must consist of 20 stations. All surveyors are asked to mark locations of their survey stations on 1:50,000 scale topographic maps and to complete a form describing the location and general habitat of each stop. Surveyors taking over an existing route are encouraged to use the same stops as the previous surveyor.

All routes are surveyed within 30 minutes after sunrise. Recommended weather conditions for the survey are: a minimum of 1.6 km visibility, wind speed less than 13 km/h, and no persistent rain or snow. If the weather changes or exceeds any of these limitations during the survey, surveyors are asked to use their best judgement as to whether to complete the survey or stop and

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 5 of 25 return at a later date to re-do the survey. Surveyors record time and weather conditions at the beginning and end of each survey route.

At each station, a pre-recorded cassette tape or CD consisting of six sets (20 seconds each) of Red-shouldered Hawk calls interspersed with 40 seconds of silence is played. Surveyors place the broadcast unit on a towel on the roof of their vehicle with the speaker facing one side of the road. The speaker is rotated 180° after each broadcast call interval, so that three sets of calls are broadcast to each side of the road. After the 5:20 minute broadcast period, the surveyor remains at the station for a final 2-minute listening period before moving to the next station.

When surveying a route, a single surveyor usually makes all observations, but a second person often accompanies the surveyor to record data. Numbers of adult, immature, and unknown-age raptors seen or heard, and the numbers of woodpeckers seen, heard calling, or heard drumming are recorded at each stop. Volunteers are also asked to record additional noteworthy observations, especially VTE (Vulnerable, Threatened, Endangered) or rare birds. These additional observations are entered separately into the Ontario Birds at Risk (OBAR) database.

While developing the protocol, there were some changes in the survey. The duration of the listening period and the volume of the tape both changed substantially after 1990, so data from 1990 are excluded from analyses. Also, the same broadcast tape was used from 1991 to 1994, and then new tapes were produced in 1995. Although these new tapes contained the same recordings, they were louder and of better quality, especially because the old tapes had deteriorated during the four years that they were used. The same tapes were again used between 1995-1997. Starting in 1998, new tapes were distributed annually to prevent deterioration. Although we have tried to keep the volume and quality of the tapes the same as those released in 1995, there has been some annual variation in tape quality.

Throughout the duration of the survey, there have been several changes in the collection of woodpecker data. Starting in 1992, participants were asked to record only the two target species (Pileated Woodpecker and Yellow-bellied Sapsucker). Participants were asked to record all woodpecker species starting in 1995. In 1996, a 30-minute training tape with examples of hawk calls and woodpecker calls and drumming was distributed to help surveyors identify hawk and woodpecker species. The data form was also redesigned in 1996 to encourage participants to report all woodpecker species.

In 1995, we asked surveyors to record whether they were able to identify the two target woodpecker species by sight, call, or drumming, and in 1996, volunteers were asked to provide the same information for all woodpecker species. Most participants are able to identify all species by sight, but many are not confident at identifying one or more species (including the target species) by calls or drumming.

In this report, trend analyses for Pileated Woodpecker and Yellow-bellied Sapsucker are restricted to surveys for which the surveyor was able to identify the species by call, sight and

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 6 of 25

drumming. For all other woodpecker species, birds recorded based only on drumming alone were excluded because many surveyors could not identify drumming with confidence. This approach allowed inclusion of data from all surveyors able to identify each species by call and by sight. For the two target species, data from 1992-2003 are included in trend analyses. However, prior to 1995, data were included from all surveyors regardless of identification skills. For the three other species, only data from 1996-2003 were included in trend analyses.

BROADCAST EQUIPMENT

In the past, BSC provided volunteer surveyors with one of two types of broadcast equipment:

(1) Car stereo set. Yamaha YCR-340 or YCR-325 car cassette decks (16 units) with Yamaha- 6920 bookshelf speakers or a horn speaker, or (2) Portable cassette player. Sanyo MCD-S750 (29 units) or Fisher PH-W3300 (1 unit) portable CD/cassette player equipped with a horn speaker in place of the manufacturer's speakers.

Some participants are still using these units, but many of the units are no longer functioning adequately, so are no longer being used. Most participants are now using their own equipment. Because different equipment may broadcast at different power levels, this may be affecting hawk responses. Standardization of broadcast equipment would be the best solution, but the costs of providing broadcast units to all surveyors are prohibitive.

ROUTE SELECTION

The number of routes increased slowly from 1990 to 1993 and underwent a major expansion (37 new routes) in 1994. The initial set of routes was selected either by the survey coordinator or the participants in habitat known to be suitable for Red-shouldered Hawks, although hawks were not necessarily known to occur along the route itself. When the survey was expanded in 1994, a stratified random route selection protocol was introduced, based upon map grid squares. Data from the first Ontario Breeding Bird Atlas (Cadman et al. 1987) and the Ontario Rare Breeding Bird Program (Austen et al. 1994) were used to determine the number of 10 km x 10 km atlas squares (defined by the Universal Transverse Mercator (UTM) grid) in each 100 km x 100 km block that have had breeding Red-shouldered Hawks reported since 1981. The number of routes per block was assigned proportional to the number of squares with hawk records in that block. The starting point for each route was placed in a square that was randomly chosen from among those known to contain Red-shouldered Hawks in that block. The actual route location was then determined based upon the availability of suitable secondary roads that were accessible in early spring. The route did not necessarily remain in the chosen square, although it was required to remain at least 3 km from any other route. Not all randomly chosen routes were surveyed because of the absence of volunteers in certain locations. Between 1995-2001, a number of new routes were established in locations selected largely by the surveyors.

To ensure that trends observed along selected routes were representative of those in the region as

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 7 of 25

a whole, a route randomization strategy was adopted for all new routes in 2002. To do so, we first divided the province into blocks, where each block corresponded to a 1:250,000 topographic map. The blocks were then subdivided into east and west such that each block measured 1° latitude x 1° longitude. Blocks south of 43° latitude or north of 47° latitude were excluded because they were outside of the Red-shouldered Hawk’s breeding range in Ontario. Blocks that were less than 1/3 covered by land or less than 1/3 within political boundaries of Ontario were also excluded.

To identify blocks with insufficient coverage, the starting points for all existing routes were overlaid on the grid. Minimum coverage was set at four routes per degree block if the block was more than 50% covered by land and two routes/block if there was 33-50% land coverage. After analyzing the current route distribution, routes were then selected randomly in blocks that had insufficient coverage. To select the routes, the blocks were divided into 16 cells (1 cell=1:50,000 topographic map) and a random cell (or cells) was selected from the block. If the cell that was randomly selected had no suitable habitat or suitable roads, then another cell was randomly selected until an appropriate cell was identified. In some cases, entire blocks were eliminated because the habitat was completely unsuitable. Once the cells were selected, a random point was selected within that cell and the road nearest to the random point was selected as the starting point.

A total of 19 random starting points were selected using this methodology. Volunteers that were assigned to random routes were asked to scout the area ahead of time to ensure that the habitat and road were suitable. Nine random routes were assigned to surveyors in 2003, but only one was run, and two were scouted to determine if the area was suitable for Red-shouldered Hawks.

DATABASE STRUCTURE

The Red-shouldered Hawk database was previously maintained in Paradox but was converted into Microsoft Access in 2002 to be compatible with the scannable data forms. The main database structure is comprised of 11 tables, each prefixed with rsha_. The contents of the various database tables, pertinent queries, and relationships are summarized below.

Contents of database tables

1. route: basic route information including route number, route name, nearest town, modification, current status, and current surveyor information 2. modif_routes: information pertaining to what year route modifications occurred 3. surveyors: contact information for past and present surveyors. 4. surveys: survey information describing date surveyed, start and end times, weather conditions, surveyor ID 5. stations: station specific data including an odometer reading of distance to each station, and the time each station was surveyed 6. coordinates: geographic information for each route: closest town, station specific coordinates

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 8 of 25

(UTM, Lat/Long, Map Number), and written stop descriptions provided by surveyors 7. id_wp: information for each route/year combination on whether the surveyor was unable to identify any of the woodpecker species based on categories of ‘Call’, ‘Sight’, or ‘Drumming’ These data are used to identify routes with surveyors who were confident in their identifications. This information was only available starting in 1995 for Pileated Woodpecker and Yellow- bellied Sapsucker, and 1996 onwards for remaining species. 8: equipment_test: annual information describing type and sound quality of surveyor broadcast equipment 9. data_hawk: hawk observations by station including species and numbers in each age class (adult, immature or unknown). Unidentified Accipiters and unidentified Buteos are also recorded. 10. data_wp: woodpecker observations by station including species, total number of individuals, and those detected only by drumming. Unidentified Woodpeckers are also recorded. 11. waitinglist: list of names, addresses, and contact information for volunteers interested in participating

Queries

The “Yearly_routes_per_constituent” query allows the user to view the route and surveyor history. This cross tabulation query displays how many times a route was surveyed in each year and by whom. This query is very useful when a volunteer calls with questions about his/her route because it allows the user to view the entire route history while on the phone.

Relationships

“Surveys” is the main header table in the database and is linked to four tables, each containing survey specific information: Stations, Route, Equipment_test, and Id_wp. “Stations” is further linked to “Data_hawk” & “Data_rsha”, while "Route" is further linked to "Coordinates" and "Modif_routes".

Referential integrity is enforced to ensure that all relationships between related tables are valid. This means that all records in the primary table must also be found in the related table. For example, information contained in the “Stations” table must also be found in “Surveys”.

DATA ANALYSIS

In this report, I used Poisson regression models of Link and Sauer (1997, 1998), but did not adjust for any potential over-dispersion in the data relative to a Poisson model. These models were implemented with PROC GENMOD in SAS. This approach was used to provide annual indices for Red-shouldered Hawks and the five most commonly encountered woodpecker species. It was also used to estimate long-term trends for Red-shouldered Hawks, assuming that the population has been changing by the same proportion every year (i.e. exponentially). Data from 1991 to 2003 were included in these analyses.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 9 of 25

RESULTS

VOLUNTEER PARTICIPATION

In 2003, 98 routes were assigned to volunteers, including 2 new routes (1 random, 1 non- random). Although 98 routes were assigned, 38 (39%) were not surveyed, or the data were not returned to us. Since 2000, the proportion of volunteers that failed to complete their assigned route has been increasing (2000: 15%, 2001: 31%, 2002: 38%). In 2003, 53 volunteers surveyed a total of 60 routes, 50 survey pairs ran a single route, 1 pair surveyed 2 routes each, 1 pair surveyed three routes and one pair surveyed 5 routes. There were 55 routes that had 18 completed stops, which were used for analyses. The locations of all routes that have been run since the beginning of the survey are shown in Figure 1.

SURVEY CONDITIONS

In 2003, seven routes were surveyed outside of the recommended time period (17 April - 7 May). All seven were surveyed after 7 May. All routes except one were surveyed at the appropriate time of day (within one half hour of sunrise). Starting temperatures for survey routes ranged from –5º C to 15ºC, and ending temperature varied from 1ºC to 22ºC. All surveys were initiated with wind conditions four or less on the Beaufort scale, and with little or no precipitation. Cloud cover varied considerably among surveys.

SURVEY RESULTS

Results presented here include only surveys for which at least 18 stops were completed (N=55 for 2003). Data from all surveys, regardless of duration or quality, have been stored in our data base, and may potentially be included in future analyses. For this report, we have not excluded routes on the basis of inappropriate survey dates or time of day, but such screening may be considered in the future.

RED-SHOULDERED HAWKS

In 2003, 167 Red-shouldered Hawks were reported along 37 of the 55 completed survey routes (67%), for a mean of 3.04 hawks per route (Table 1). Compared to last year, a smaller proportion of routes recorded Red-shouldered Hawks (2002: 73%) and mean hawks per route was also lower (2002: 3.22). Mean number of Red-shouldered Hawks recorded this year was the second lowest since the survey began. The distribution of Red-shouldered Hawks in the study area is shown in Figure 2.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 10 of 25

Figure 1 – Distribution and status of Red-shouldered Hawk and Spring Woodpecker survey routes in 2003.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 11 of 25

Figure 2 - Mean number of Red-shouldered Hawks recorded by participants in the Red-shouldered Hawk and Spring Woodpecker Survey from 1991-2003.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 12 of 25

To estimate how numbers in 2003 compared with previous years, the Poisson modelling approach of Link and Sauer (1997, 1998) was used to calculate annual indices controlling for differences among routes. The number of Red-shouldered Hawks reported per route varied considerably among years, although there has been little change from 1998-2003 (Figure 3). The number of hawks declined gradually from 1991 to 1994, with a statistically significant decline between 1993-1994. This was followed by a significant increase in 1995. From 1995-1997, population indices declined again, which was followed by a significant increase in 1998. The observed increases in detection of Red-shouldered Hawks in 1995 and 1998 may be due to increases in tape quality, as new tapes were issued in both of these years (Francis 1999). There was no significant change in the population index between 1998 and 2002. Between 2002 and 2003, the number of Red-shouldered Hawks detected declined significantly (P<0.10).

To estimate if there has been any average long-term change in Red-shouldered Hawk indices, I used a Poisson model, with routes as covariates, treating year as a continuous variable. This analysis showed no significant change in the Red-shouldered Hawk population from 1991-2003 (slope of +0.8% per year, 95% Confidence Interval: -1.3 to 1.5%, P = 0.9). However, models assuming linear change are not necessarily appropriate because of the large annual fluctuation (Figure 3).

7.0

6.0

5.0 *

+ 4.0 ** * +

3.0 + 2.0

1.0

0.0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Figure 3 – Estimated annual indices for Red-shouldered Hawk counts from 1991-2003 derived from a generalized linear model assuming Poisson residuals and a log-link function. 95% confidence limits refer to differences from 2003, which was chosen as the baseline year. Year-to-year comparisons based on post-hoc contrasts were significantly different as marked: + P<0.10, * P<0.05, ** P<0.01.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 13 of 25

OTHER RAPTORS

Fourteen other raptor species have been recorded during the Red-shouldered Hawk surveys (Table 1), including two additional raptor species at risk. In 2001 and 2002, a Peregrine Falcon was observed on the survey and in 2002 Bald Eagle was added to the species list. Red- shouldered Hawk was encountered more frequently than any other raptor species, presumably reflecting the choice of habitat for conducting the survey, as well as the use of a broadcast tape to increase response of the target species. Turkey Vulture was the second most common raptor species recorded by volunteers (mean: 1.38 birds/route), followed by Broad-winged Hawk (mean: 1.07 birds/route) and Red-tailed Hawk (mean: 0.22 birds/route). Average numbers of each raptor species do not appear to have changed dramatically between 1991 and 2003 (Table 1).

WOODPECKERS

In addition to the two target woodpecker species (Pileated Woodpecker and Yellow-bellied Sapsucker), three other woodpecker species were commonly recorded during 2003 surveys: Downy Woodpecker, Hairy Woodpecker, and Northern Flicker (Table 2). Red-bellied Woodpeckers were reported in 1997, 1999, 2002-2003 but were not recorded in 2000 or 2001. Three Red-headed Woodpeckers were recorded in 1997, 1999, and 2000; two were recorded in 2002, but none were recorded in 2001 or 2003.

The most abundant woodpecker species was Yellow-bellied Sapsucker, followed by Northern Flicker (Table 2). Both of these species may still be migrating at the time of the survey, and hence may be at higher densities than they would be during the breeding season. Totals probably also include residents already settled on territories. The Hairy Woodpecker, Pileated Woodpecker, and Downy Woodpecker are year-round residents that do not usually migrate. Hence, all three of these species were presumably settled on breeding territories at the time of the survey. All five woodpecker species were found throughout the study area, but the areas of highest densities were not necessarily the same for all species (Figure 4).

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 14 of 25

Table 1 - Mean number of raptors per route on Red-shouldered Hawk and Spring Woodpecker surveys from 1991-2003. Only routes with at least 18 stops were included.

Species 91 92 93 94 95 96 97 98 99 00 01 02 03 Mean American 0.27 0.31 0.24 0.25 0.08 0.22 0.15 0.14 0.15 0.24 0.23 0.42 0.18 0.22 Kestrel Broad-winged 0.93 1.25 1.12 0.57 0.48 0.56 0.58 0.74 0.75 0.93 1.21 1.07 1.07 0.87 Hawk Cooper’s Hawk . . 0.12 0.04 0.10 0.07 0.05 . 0.04 0.02 0.06 0.04 0.04 0.06 Merlin . . . 0.02 0.02 0.07 0.03 0.07 0.10 0.06 0.02 0.12 0.07 0.06 Northern 0.07 . . . 0.08 0.15 0.05 0.02 0.02 0.04 0.06 0.04 0.02 0.05 Goshawk Northern Harrier 0.47 0.31 0.06 0.25 0.25 0.29 0.08 0.21 0.25 0.13 0.10 0.28 0.15 0.22 Osprey 0.07 0.31 0.18 0.18 0.02 0.34 0.08 0.29 0.13 0.35 0.15 0.19 0.18 0.19 Rough-legged 0.07 . . . 0.06 . . . . 0.06 0.06 0.16 0.04 0.07 Hawk Red-shouldered 5.13 3.38 3.94 2.24 4.69 4.76 4.08 4.62 3.96 3.96 3.50 3.23 3.04 3.89 Hawk Red-tailed Hawk 0.33 0.81 0.12 0.31 0.21 0.29 0.38 0.10 0.46 0.28 0.24 0.12 0.22 0.30 Sharp-shinned 0.27 0.13 . 0.12 0.10 0.22 0.25 0.12 0.04 0.11 0.11 0.05 0.04 0.13 Hawk Turkey Vulture 3.00 1.81 1.82 1.18 1.56 2.66 1.98 2.33 2.12 2.44 2.21 1.86 1.38 2.03 Unknown 0.07 0.13 0.12 0.04 . 0.12 0.05 0.02 0.04 0.04 0.02 . 0.04 0.06 Accipiters Unknown 0.33 0.13 0.59 0.22 0.08 0.12 0.10 0.05 0.21 0.06 0.23 0.07 0.07 0.17 Buteos Number of 15 16 17 51 52 41 40 40 52 54 62 57 55 routes

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 15 of 25

Table 2 – Number of each woodpecker species recorded on 2003 Red-shouldered Hawk and Spring Woodpecker Survey (N=55 routes), as well as the number of routes that recorded each species and the mean number of birds per route. Drum Only is the number of woodpeckers recorded on the survey that were identified by drumming alone.

Species Birds Routes Mean per route Drum Only Downy Woodpecker 219 39 4.0 99 Hairy Woodpecker 200 45 3.6 91 Northern Flicker 374 50 6.8 35 Pileated Woodpecker 141 39 2.6 48 Yellow-bellied Sapsucker 702 46 12.8 439 Red-headed Woodpecker 0 0 0 0 Red-bellied Woodpecker 1 1 0.02 0 Unknown Woodpecker 106 18 1.9 96

To estimate temporal change in woodpecker numbers, I calculated annual indices for each species using the same approach as for Red-shouldered Hawks (i.e., Poisson regression, controlling for differences among routes). For Pileated Woodpecker and Yellow-bellied Sapsucker, data from 1992-2003 were used, and for the three other woodpecker species data from 1996-2003 were used. Data from routes on which the observer could not confidently identify the species based on both calls and sight were excluded from the trend analyses.

The indices appear to have changed over time for all species, although population indices for Northern Flicker have been the most stable (Figure 5). Population indices for Pileated Woodpecker and Yellow-bellied Sapsucker have shown the largest annual fluctuations of the five woodpecker species. Population indices for Pileated Woodpecker increased between 1994 and 1996 and reached their highest levels in 1997, 1999 and 2003. There were significant declines in 1998 and 2000 and no change in population indices between 2000 and 2002. Yellow- bellied Sapsuckers steadily increased between 1993 and 1995, declined between 1995 and 1997, and then reached their highest point in 1999. Similar to Pileated Woodpeckers, Yellow-bellied Sapsuckers also showed a significant decline between 1999 and 2000. This species showed a population increase in 2002 and no change between 2002 and 2003.

Downy and Hairy Woodpeckers have shown very similar population trends from 1996-2003 (Figure 5). Both species increased steadily from 1996 to 1998 and have leveled off since 1998. Northern Flicker population indices have been fairly stable over the duration of the survey. They showed a significant increase between 1998 and 1999 and a significant decline in 2003.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 16 of 25

Figure 4 - Mean number of Pileated Woodpecker, Yellow-bellied Sapsucker, Hairy Woodpecker, Downy Woodpecker, and Northern Flicker recorded by participants in the Red- shouldered Hawk and Spring Woodpecker Survey from 1995-2003.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 17 of 25

Figure 5 – Annual population indices for five woodpecker species detected regularly during the survey. Indices were calculated using generalized linear regression with Poisson residuals and a log-link function. 95% confidence limits of estimates refer to deviations from the 2003 value. Observations from surveyors who were not confident in their identifications were excluded from analyses. Year-to-year comparisons based on post-hoc contrasts were significantly different as marked: + P<0.10, * P<0.05, ** P<0.01.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 18 of 25

Figure 6– Annual population indices for five woodpecker species regularly encountered on Red-shouldered Hawk and Spring Woodpecker surveys (1996-2003) and population indices derived from Breeding Bird Survey routes (1996-2002) in Ontario. Solid lines with dots are indices from Red-shouldered Hawk and Spring Woodpecker Survey, dashed lines with solid squares are indices from Breeding Bird Survey.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 19 of 25

WOODPECKER TRENDS FROM THE BREEDING BIRD SURVEY

Population indices for Pileated Woodpecker, Yellow-bellied Sapsucker, Downy and Hairy woodpeckers, and Northern Flicker derived from Breeding Bird Survey (BBS) routes (1992-2002) in Ontario were plotted with indices derived from this survey (Figure 6). Pileated Woodpecker population indices from BBS are quite similar to those derived from the Red-shouldered Hawk and Spring Woodpecker Survey, especially for the years 1996- 2002. Patterns in Northern Flicker population indices also show good agreement between the two surveys. Trends in Downy Woodpecker population indices are similar between BBS and this survey, with the exception of 2001-2002 when opposite patterns were observed. Unlike the other woodpecker species, population indices of Yellow-bellied Sapsuckers and Hairy Woodpeckers showed little concordance between the two surveys.

DISCUSSION

VOLUNTEER PARTICIPATION

Over the past three years, the proportion of volunteers that do not complete assigned routes has increased. Each year, we ensure that as many routes as possible are assigned to volunteers (usually close to 100%). In addition, volunteers are urged to complete their route as assigned. If the assigned surveyor is unable to complete his/her assigned route, we do not attempt to find a new volunteer to fill in for a single season because it is important to have the same surveyor assigned to a route for as many years as possible. To increase participation, we will be establishing 5 new routes in 2004, and will be contacting volunteers that have not participated for three years to determine whether a new surveyor should be assigned, and to determine why volunteers drop out of the program or do not complete their assigned routes.

RED-SHOULDERED HAWK POPULATION CHANGES

The number of Red-shouldered Hawks detected on survey routes declined significantly between 2002 and 2003 (P<0.10) which is the first observed decline since 1996. In addition, mean number of Red-shouldered Hawks detected per route was the second lowest since the survey began. It is not known why fewer Red-shouldered Hawks were detected during the 2003 survey. Mean survey dates and weather conditions during 2003 surveys were comparable to those of other years. Although average weather conditions during surveys were not unusual, spring temperatures in 2003 were lower than average, which may have affected the breeding chronology of Red-shouldered Hawks.

Despite this small drop in numbers in 2003, the Red-shouldered Hawk population in central Ontario has been stable with fairly minor annual fluctuations from 1991-2003.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 20 of 25

This suggests that although the population remains small and is considerably depressed relative to its historic abundance, there is sufficient habitat in central Ontario to maintain the present Red-shouldered Hawk population. However, the continued expansion of urban areas and ever-increasing cottage development in prime Red-shouldered Hawk breeding habitat are ongoing threats to the Ontario population.

Data from both this survey and the first three years of the 2nd Ontario Breeding Bird Atlas suggest that the Red-shouldered Hawk has expanded its range northward in Ontario. Breeding Bird Atlas data from 2001-2003 showed that there are many 10 km x 10 km atlas squares in southwestern Ontario that reported Red-shouldered Hawks in the first atlas (1981-1985) but have failed to do so in the second atlas. The opposite is true for central Ontario where Red-shouldered Hawks have been discovered in many squares in the second atlas but not in the first. These data suggest that Red-shouldered Hawks may have been forced farther north in Ontario to locate appropriate breeding habitat.

WOODPECKER POPULATION CHANGES

Pileated Woodpeckers increased significantly during the 2003 survey and Northern Flickers declined. All other woodpecker species showed no change this year. A comparison of trends in population indices for woodpeckers from this survey and the Breeding Bird Survey show good agreement for some species, but not for others. Interestingly, Pileated Woodpecker indices from BBS show similar trends to indices from this survey for the period 1996-2002. In 1996, volunteers were provided with a training tape for hawks and woodpeckers for the first time, and the data form was redesigned to encourage volunteers to record woodpecker data. These changes in protocol may have resulted in improved monitoring of some woodpecker species.

Yellow-bellied Sapsuckers and Northern Flickers are species that may still be migrating at the time of the survey, so counts include both migrants and residents. The good agreement between BBS and the Red-shouldered Hawk survey for Northern Flicker suggest that this survey is tracking trends in Northern Flicker populations. However, the opposite is true for Yellow-bellied Sapsuckers where there was no agreement between the two surveys. The Breeding Bird Survey is more likely to adequately monitor Northern Flickers than Yellow-bellied Sapsuckers because flickers are more visible (i.e. more vocal and active) during June, when BBS routes are run, than are sapsuckers. Very few Yellow-bellied Sapsuckers are detected on BBS routes in Ontario.

ACKNOWLEDGMENTS

Many, many thanks to all the dedicated volunteers who collected the 2003 data used in this report. We hope they will all continue to participate in this valuable project in the future. Thanks to Margaret McLaren for helping to develop this program and to Jon McCracken

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 21 of 25 for project management and editorial assistance. Thanks to Susan Debreceni for helping with volunteer recruitment and coordination. Red-shouldered Hawk illustrations in the report are by Peter Burke; Charles Francis wrote the original SAS programs used to analyze data. Funding for this project was provided by the Wildlife Assessment Program of the Ontario Ministry of Natural Resources.

LITERATURE CITED

Austen, M.A. 1991. Testing a survey technique for monitoring relative abundance of Red- shouldered Hawks in Ontario. Unpublished report by Federation of Ontario Naturalists for Ontario Ministry of Natural Resources.

Austen, M.A, and M.D. Cadman. 1994. The status of the Red-shouldered Hawk (Buteo lineatus) in Ontario. Unpublished report by Ontario Rare Breeding Bird Program, Federation of Ontario Naturalists for Ontario Ministry of Natural Resources, Terrestrial Ecosystems Branch, Toronto, Ontario.

Austen, M.A., M.D. Cadman, and R.D. James. 1994. Ontario Birds at Risk: status and conservation needs. Federation of Ontario Naturalists, Don Mills, Ontario and Long Point Bird Observatory, Port Rowan, Ontario.

Bryant, A.A. 1986. Influence of selective logging on Red-shouldered Hawks, Buteo lineatus, in Waterloo Region, Ontario, 1953-1978. Canadian Field Naturalist 100:520-525.

Cadman, M. D., P. F. J. Eagles, and F. M. Helleiner. 1987. Atlas of the breeding birds of Ontario. Federation of Ontario Naturalists and Long Point Bird Observatory, University of Waterloo Press.

Callaghan, C. 1990. Testing a survey technique for monitoring relative abundance of red- shouldered hawks in Ontario. Unpublished report by Ontario Rare Breeding Bird Program, Federation of Ontario Naturalists for Ontario Ministry of Natural Resources.

Crocoll, S. 1994. Red-shouldered Hawk (Buteo lineatus). In the Birds of North America, No. 107. (A.Poole and F. Gill, eds.). The Birds of North America Inc., Philadelphia, PA.

James, R.D. 1991. Annotated checklist of the birds of Ontario. Royal Ontario Museum, Toronto.

Link, W. A., and J. R. Sauer. 1997. Estimation of population trajectories from count data. Biometrics 53: 488–497.

Link, W. A., and J. R. Sauer. 1998. Estimating population change from count data:

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 22 of 25 application to the North American Breeding Bird Survey. Ecological Applications 8: 258– 268.

McLaren, M. 1993. The road-based transect program for monitoring Red-shouldered Hawk populations in Ontario: a strategic review. Unpublished report, Ontario Ministry of Natural Resources, Terrestrial Ecosystems Branch, Toronto, Ontario.

Peterjohn, B. G., J. R. Sauer, and W. A. Link. 1994. The 1992 and 1993 summary of the North American Breeding Bird Survey. Bird Populations 2: 46-61.

Risley, C.J. 1982. The status of the Red-shouldered Hawk in Ontario, with an overview of the status in Canada. Unpublished report, Non-Game Program, Wildlife Branch, Ontario Ministry of Natural Resources, Toronto, Ontario.

Sutherland, D.A. (comp.) 1994. Natural Heritage Resources of Ontario: Birds. Unpublished report by Natural Heritage Information Centre, Peterborough, Ontario. 22 pp.

Szuba, K.J. 1990. Proposed monitoring scheme for Red-shouldered Hawks and Cooper's Hawks. Unpublished report Wildlife Policy Branch, Ontario Ministry of Natural Resources, Toronto.

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 23 of 25

APPENDIX A - Route number, name, latitude and longitude for all routes that have been run at least once, with the numbers of hawks detected in each year. AB indicates that the route has been abandoned; A indicates active routes, I indicates inactive routes. Cells containing ‘.’ indicate that the route was not run (or the data were not reported) in that year. Route # Route Name Status Lat. Long. Number of RSHA detected by year 91 92 93 94 95 96 97 98 99 00 01 02 03 1 Sparrow Lake A 44 48' 79 23' 14 . . 4 17 5 2 2 6 5 3 . 2 2 Cashel A 44 58' 77 29' 6 2 0 1 5 2 . 0 2 2 5 . 4 3 Ashdad AB 45 21' 76 39' 0 0 0 ...... 4 Canoe Lake A 44 30' 76 35' 22 21 20 22 30 33 29 29 35 32 36 34 31 5 Big Cedar Lake A 44 35' 78 08' 2 2 0 0 2 . 1 1 2 2 3 0 1 6 Livingstone Lake A 45 17' 78 52' 0 2 0 1 . . . 2 7 0 0 0 1 7 Foreman A 45 07' 79 35' 8 13 9 1 10 5 4 7 . 3 1 4 5 8 White Lake A 45 16' 76 20' 6 2 2 3 4 6 5 4 4 3 4 3 2 9 Lavant A 45 02' 76 45' 1 1 0 1 0 . . . 4 5 3 1 0 10 Opinicon Lake A 44 35' 76 19' 10 5 15 11 12 10 9 20 20 23 22 18 27 11 Brinkmans Corners A 45 11' 81 24' 3 0 0 0 1 0 1 . 0 1 1 2 0 12 Quadeville A 45 19' 77 23' 0 1 0 0 1 . . 1 . . 0 . 0 13 Mountsberg I 43 27' 80 01' 0 0 0 0 . . 1 ...... 14 Kennisis Lake AB 45 08' 78 37' 0 0 . 0 0 ...... 15 Maple A 44 00' 79 22' 5 4 4 0 4 4 2 1 5 8 1 1 5 16 Brown Hill A 44 15' 79 22' . 1 0 0 2 2 4 . 0 1 . . 2 17 Norfolk I 42 42' 80 27' . 0 1 0 0 . . 2 1 1 0 0 . 18 Minden A 44 59' 78 45' . . 11 6 13 3 . 2 . 0 . 0 0 19 Belmont-Methuen A 44 34' 78 02' . . 5 . 4 0 2 4 4 9 3 2 2 20 Parham A 44 39' 76 45' . . . 1 0 4 1 3 3 1 0 . 2 21 Ardoch I 44 52' 76 45' . . . 3 9 7 4 6 4 2 2 . . 22 Sharbot Lake A 44 46' 76 41' . . . 3 9 11 . . . 5 3 . 8 23 Myers Cave A 45 11' 81 24' . . . 4 16 . 9 . 4 . . . 0 24 Steenburg Lake A 44 50' 77 50' . . . 5 5 6 . 10 4 12 5 . 1 25 Vansickle A 44 30' 77 42' . . . 3 5 . . 3 3 0 4 7 5 26 Otty Lake I 44 50' 76 12' . . . 12 21 11 10 17 18 10 . . . 27 Zephyr A 44 11' 79 15' . . . 2 1 0 . 3 . 1 . 2 2 28 Uffington A 44 59' 79 11' . . . 0 2 1 0 3 0 0 3 1 0 29 Union Creek I 44 43' 78 37' . . . 1 . . . . . 0 . . . 30 Farlain Lake A 44 50' 79 54' . . . 5 7 2 0 2 1 . 3 3 7 31 Cavan A 44 12' 78 28' . . . . 3 0 . . 1 1 1 2 1 32 Ifracombe AB 45 21' 79 16' . . . 1 ...... 33 Beatrice I 45 17' 79 18' . . . . 0 0 . . 0 . 0 0 . 34 Fraserburg I 45 03' 79 15' . . . 3 5 . . . 0 . 1 . . 35 Humphrey I 45 15' 79 49' . . . 1 . . . . . 3 . . . 36 Broadbent AB 45 26' 79 41' . . . 0 ...... 37 Twelve Mile Bay A 45 04' 79 48' . . . 1 . 7 . . 0 3 0 1 2 38 Harburn AB 45 04' 78 31' . . . 0 ...... 39 Flower Station A 45 09' 76 41' . . . 0 4 . . 1 0 . . . 1

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 24 of 25

Appendix A continued. Route # Route Name Status Lat. Long. Number of RSHA detected by year 91 92 93 94 95 96 97 98 99 00 01 02 03 40 Esmonde AB 45 23' 77 01' . . . 1 ...... 41 Manion Corners AB 45 18' 76 02' . . . 0 ...... 42 Lemieux A 45 21' 75 08' . . . . 0 ...... 0 0 43 Glenroy I 45 11' 74 39' . . . 0 8 . 3 ...... 44 Charles Lake I 44 43' 81 02' . . . 0 2 1 1 1 . . . . . 45 Waubamik A 45 39' 80 02' . . . 2 3 . 1 2 1 0 2 . 2 46 Commanda A 45 53' 79 46' . . . 4 1 1 3 . 1 7 1 2 2 47 Pontypool AB 44 06' 78 38' . . . 0 0 0 . . 0 1 0 . . 48 Kinlough I 44 04' 81 26' . . . 0 2 0 0 0 0 0 . . . 49 Beverly Swamp A 43 24' 80 06' . . . 0 1 1 0 0 0 . . 0 0 50 Wrigley's Corners AB 43 20' 80 26' . . . 0 . . . 0 0 . . . . 51 St. Joseph Island W A 46 19' 84 07' . . . 2 5 7 3 . 7 5 11 8 2 52 Mountain Lake I 44 41' 81 01' . . . 2 4 6 5 3 2 1 1 . . 53 Mooresburg A 44 20' 80 58' . . . 2 2 6 3 5 2 0 4 3 4 54 Burke Settlement I 44 51' 76 44' . . . 3 6 6 7 ...... 55 Buckhorn I 44 33' 78 21' . . . 3 1 4 ...... 56 Greenfield A 45 18' 74 44' . . . 0 2 ...... 1 0 57 Petawawa AB 45 59' 77 25' . . . . 0 ...... 58 Catchacoma I 44 47' 78 22' ...... 2 3 . . . 59 Pine Springs I 45 08' 78 51' ...... 1 . . 60 Fox Lake Road A 46 09' 81 44' . . . . 0 0 0 0 2 1 0 4 2 61 Two Island Lake Roa I 45 19' 77 02' . . . . 0 0 . 2 . . 0 . . 62 Burnstown A 45 26' 76 38' . . . . 9 7 8 10 8 3 5 4 1 63 Koshkawong A 46 14' 83 59' . . . . 5 8 9 . 10 7 12 7 7 64 Sauble- Howdenvale A 44 40' 81 15' . . . . 0 7 5 9 7 7 8 9 4 65 Limoges AB 45 21' 75 13' . . . . 1 ...... 66 Ingleside A 44 58' 75 02' . . . . 0 . 0 . 0 . 0 . 0 68 Dog Lake I 44 28' 76 19' ...... 7 4 7 5 . . . 69 Christie Lake A 44 48' 76 25' . . . . . 17 16 12 11 7 12 15 9 70 Glen Alda I 44 50' 77 55' . . . . . 0 0 ...... 71 Mountain Grove I 44 44' 76 51' . . . . . 5 . . 6 5 4 . . 73 Wasi Lake Circle I 46 11' 79 19' . . . . . 0 0 . 0 . . . . 74 Big Chute A 44 48' 79 37' ...... 3 4 2 4 5 3 8 76 Renderville Road AB 44 04' 77 33' ...... 0 . . . . 0 . 78 Killbear I 45 22' 80 16' ...... 3 9 7 7 . 5 . 79 Hilton Falls A 43 30' 80 00' ...... 2 1 0 1 0 0 0 80 Trout Lake I 46 18' 79 22' ...... 0 0 . . . . 81 Red Rock Road AB 46 36' 84 33' ...... 1 0 0 . . . 82 Echo Bay A 46 32' 84 06' ...... 4 2 0 6 3 0 83 Laird A 46 23' 84 07' ...... 4 1 2 5 3 2

2003 Red-shouldered Hawk and Spring Woodpecker Survey Page 25 of 25

Appendix A continued. Route # Route Name Status Lat. Long. Number of RSHA detected by year 91 92 93 94 95 96 97 98 99 00 01 02 03 84 Robertson Lake I 46 46' 84 16' ...... 0 1 1 . 85 Gordon Lake A 46 26' 83 55' ...... 5 5 4 2 86 Limberlost Rd I 45 26' 79 05' ...... 0 . . 87 Torrance-Southwood A 44 59' 79 34' ...... 6 3 3 4 88 Whitefish-Cartier A 46 42' 81 34' ...... 2 0 1 0 90 Nipissing Road I 47 08' 79 46' ...... 1 . . . 91 Trout Creek I 45 58' 79 18' ...... 1 0 0 . 94 Sydenham I 44 28' 76 36' ...... 2 . 95 Burnt River A 44 42' 78 43' ...... 3 7 2 96 Moonstone A 44 41' 79 41' ...... 5 3 0 98 Ormsby A 44 51' 77 44' ...... 0 1 0 100 Lake of Bays I 45 13' 79 07' ...... 4 3 . 101 Terra Cotta I 43 46' 80 01' ...... 0 0 . 102 Haliburton Forest A 45 14' 78 35' ...... 1 2 0 103 Eugenia/Ravenna I 44 25' 80 24' ...... 0 . . 104 Sturgeon River Vall A 44 44' 79 44' ...... 6 4 0 105 Kemptville I 45 03' 75 30' ...... 2 . . 111 Depot Lakes A 44 33' 76 44' ...... 3 0 2 112 Anstruther/Chandos I 44 44' 78 13' ...... 3 3 . 113 Carpenter Lake Rd I 46 31' 83 40' ...... 0 1 . 115 Kawartha Highlands A 44 51' 78 07' ...... 0 2 120 Massey AB 46 22' 81 12' ...... 0 . 130 Mildmay I 44 02' 80 55' ...... 0 . 132 Milverton A 43 27' 80 58' ...... 0 135 Goulais Bay I 46 45' 84 33' ...... 0 . 138 Mount Horeb A 44 16' 78 37' ...... 1 1

2003 Red-shouldered Hawk and Spring Woodpecker Survey