bp sdeis app f-1 west survey reports cape vincent

8/7/2019 BP SDEIS App F-1 WEST Survey Reports Cape Vincent

1/59

Environmental Resources Management Southwest, Inc.206 East 9th Street, Suite 1700

Austin, Texas 78701

(512) 459-4700

WEST Survey ReportsAppendix F-1

February 2011

Project No. 0092352


2/59

AVIAN AND BAT STUDIES FOR THE PROPOSED

CAPE VINCENT WIND PROJECT

JEFFERSON COUNTY, NEW YORK

Final Report

April 2006 May 2007

Prepared for:

BP Alternative Energy North America700 Louisiana Street, 33rd Floor

Houston, Texas

Prepared by:

David P. Young, Jr., Jessica J. Kerns, Christopher S. Nations, and Victoria K. Poulton

Western EcoSystems Technology, Inc.

2003 Central Avenue

Cheyenne, Wyoming 82001

November 28, 2007


3/59

Cape Vincent Wind Power Project

Avian and Bat Studies Report

WEST, Inc. i November 28, 2007

EXECUTIVE SUMMARY

BP Alternative Energy North America, Inc. (BPAE) is evaluating the feasibility of wind energy

development in Jefferson County, New York. The proposed project,Cape Vincent Wind Power

Project, is located south of the St. Lawrence River and north of Chaumont Bay, near the town of

Cape Vincent, New York. The exact location and size of the development will be based on anumber of factors including power purchase agreement(s), electricity markets, transmission

constraints, permitting, and results of site surveys.

Early project evaluation identified issues concerning potential impacts from the project on avian

and bat resources, in particular nocturnal migrant birds and migrant raptors, migrant bats, and

species of concern that may occupy the site. BPAE developed and implemented a one year avianand bat survey protocol to address the agency concerns and provide site-specific data for the

resources of concern. The study plan was reviewed and approved by the New York StateDepartment of Environmental Conservation and U.S. Fish and Wildlife Service. The primary

objectives of the study were to: provide information on avian and bat resources and use of the

study area that would be useful in evaluating potential impacts from the wind powerdevelopment, provide information on avian and bat resources and use of the study area that

would help in designing a wind project that is less likely to expose species to risk of collisions

with turbines, and provide recommendations for further studies and potential mitigation

measures, if appropriate.

The one-year avian and bat preconstruction study consisted of nocturnal marine radar sampling

during the spring and fall migration periods; diurnal point count surveys from fixed pointlocations conducive to observing raptors and other large birds; breeding bird survey point counts;

AnaBat sampling for migrating bats during the spring and fall; AnaBat sampling for resident bats

during the summer; and winter and early spring waterfowl and raptor surveys. The various studycomponents took into consideration the potential for federal and state-listed species occurrence

in the project area.

Nocturnal radar surveys were conducted most nights during the 63-day period between August

15 and October 15, 2006 and the 50-day period between April 19 and June 8, 2007. A total of

508 and 300 hours of radar sampling were conducted in the fall and spring respectively. Fall

mean and dispersion of flight direction were = 209.2 and r= 0.34 and spring mean and

dispersion of flight direction were = 34.0 and r= 0.52. The overall mean fall passage rate in

the horizontal mode was 345.8 13.3 targets/km/hr (mean SE) and the overall mean springpassage rate in the horizontal mode was 166.2 8.8 targets/km/hr (mean SE). For sampling at

the 1.5-km range in vertical mode, mean flight altitude was 490.4 1.7 m (mean SE) aboveradar level in the fall and 441.3 2.5 m arl in the spring. Approximately 7.7% of targets hadflight altitudes less than 125 m in the fall and approximately 14.0% of targets had flight altitudes

less than 125 m in the spring. Clutter from non-avian or bat targets was considered minimal;

during the fall and spring only 1% of targets were moving very slow (< 6 m/s) and not likely bird

or bat targets.

Diurnal point count surveys were conducted during the raptor migration periods in the spring and

fall 2006 and again in the spring 2007. During spring 2006, a total of 12 point count surveys


4/59



WEST, Inc. ii November 28, 2007

were conducted resulting in 777 individual birds recorded including 79 raptors of 10 species.

During the fall season, a total of 30 surveys were conducted resulting in a total of 3,050individual birds recorded including 165 individual raptors of 10 species. During the spring 2007

season, a total of 21 surveys were conducted and 1,851 individual birds were recorded including

205 individual raptors of 9 species. Canada goose was the most commonly seen bird during

spring and fall surveys. During both spring seasons, turkey vulture was the mostly commonlyrecorded raptor species (n = 29, 66.7% of surveys; n = 111, 94.4% of surveys, respectively)

followed by American kestrel (n = 13, 41.7% of surveys) in 2006 and northern harrier (n = 37,

88.9% of surveys) in 2007. In the fall, northern harrier was the most commonly recorded raptorspecies (n = 69, 76.7% of surveys), followed by turkey vulture (n = 50, 33.3% of surveys).

Other raptor species seen included: broad-winged hawk, red-tailed hawk, rough-legged hawk,

sharp-shinned hawk, osprey, peregrine falcon, and Coopers hawk. There were no spatialdifferences in raptor use across the survey points.

Point count surveys were conducted for breeding birds on June 29 and July 6, 2006. Each point

was surveyed twice, for a total of 40 survey periods. A total of 812 individual birds were

observed in 462 groups of 63 species. Red-winged blackbird, bobolink, and song sparrow werethe most common passerines observed based on mean use estimates (number observed within

400 m per 3-minute survey). Several species of interest were recorded during the breeding bird

surveys including New York state species of concern northern harrier, Henslows sparrow,

horned lark, grasshopper sparrow, and vesper sparrow, and two species on the USFWS 2002Birds of Conservation Concern list for the Lower Great Lakes/St. Lawrence Plain region,

bobolink and wood thrush.

Spring AnaBat sampling occurred between April 13 and June 2, 2006 at the project met tower

and resulted in a total of 241 bat calls recorded (4.92 calls/night) during the 49 days of sampling.

Summer sampling occurred on 15 nights between June 28 and August 8 at the met tower andrecorded a total of 431 calls (28.7 calls/night). During fall, August 13 to October 9, sampling

occurred at three different heights at the met tower. The AnaBat unit positioned at ground level

recorded the highest number of bat vocalizations per night (9.90 calls/night) over the 58 daysampling period. At least four different species of bat, eastern red bat, hoary bat, big brown bats,

and Myotis sp. were recorded during the sampling. 208 calls that were of sufficient length to

attempt species identification were submitted for quantitative analysis. Of these eastern red bat,

little brown bat, northern myotis, and Indiana bat were identified.

Winter driving surveys in the project area were conducted on nine days between November 5,

2006 and March 1, 2007. Approximately 27 hours of survey time were spent during the drivingsurveys and a total of 13.5 hours of surveys were conducted at the three fixed-point count

stations. A total of 395 individuals in 96 groups of waterbirds, waterfowl, raptors and other birdswere recorded during the winter driving surveys and 255 individuals in 87 groups were recordedduring the winter fixed point counts. Two species of waterfowl, two species of waterbird, and six

raptor species were observed either during the surveys. Canada goose was the most common

waterfowl species observed during the winter surveys based on use estimates. Rough legged

hawk and red-tailed hawk were the most common raptor species.


5/59


6/59



WEST, Inc. iv November 28, 2007

TABLE OF CONTENTS

Introduction and Background ......................................................................................................... 1

Study Area ...................................................................................................................................... 3Study Components.......................................................................................................................... 3

Nocturnal Marine Radar Survey................................................................................................. 5

Methods................................................................................................................................... 7Results..................................................................................................................................... 8

Raptor Migration Surveys......................................................................................................... 22

Methods................................................................................................................................. 22

Results................................................................................................................................... 23Breeding Bird Survey ............................................................................................................... 31

Methods................................................................................................................................. 31

Results................................................................................................................................... 33

Nocturnal AnaBat Surveys ....................................................................................................... 35Methods................................................................................................................................. 35

Results................................................................................................................................... 37Waterfowl and Winter Raptor Surveys..................................................................................... 39

Methods................................................................................................................................. 39

Results................................................................................................................................... 40Discussion..................................................................................................................................... 43

Nocturnal Marine Radar Survey............................................................................................... 43

Raptor Migration Surveys......................................................................................................... 45

Breeding Bird Survey ............................................................................................................... 47Nocturnal AnaBat Surveys ....................................................................................................... 48

Waterfowl and Winter Raptor Surveys..................................................................................... 50

References..................................................................................................................................... 50

LIST OF TABLES

Table 1. Raptors and other large bird species observed during spring and fall diurnal

raptor migration surveys at the Cape Vincent wind power project area............................25

Table 2. Flight height characteristics and exposure indices by species observed during

diurnal raptor migration surveys at the Cape Vincent wind power project area. ..............27Table 3. Avian species observed during breeding bird surveys within the Cape Vincent

wind power project area.....................................................................................................33

Table 4. Number of sampling days, total number of calls recorded, and calls/nightrecorded by each AnaBat unit at the met tower for spring, summer, and fall

sampling periods................................................................................................................38

Table 5. Relative call frequency of species recorded at the met tower during the samplingperiods of each season. ......................................................................................................38

Table 6. Number of detections by species during summer roaming AnaBat sampling................39

Table 7. Waterfowl and raptors observed while conducting winter 2007 driving surveys

at the Cape Vincent wind power project area. ...................................................................42Table 8. Waterfowl and raptors observed while conducting winter 2007 fixed point..................42


7/59



WEST, Inc. v November 28, 2007

Table 9. Results of radar studies at proposed and existing wind project sites in the U.S..............44

Table 10. Number of raptors observed per surveyor hour in the project area and at sixestablished New York spring/fall hawk watch sites. .........................................................46

Table 11. Wind projects in the U.S. with both AnaBat sampling data and mortality data

for bat species. ...................................................................................................................48

LIST OF FIGURES

Figure 1. Proposed Cape Vincent wind power project location. ....................................................2Figure 2. Land use/land cover of the Cape Vincent project area....................................................4

Figure 3. Radar sampling locations and raptor survey locations for the Cape Vincent

project area...........................................................................................................................6Figure 4. Observed flight directions at Cape Vincent project area.................................................9

Figure 5. Mean + 1 SE nightly passage rates in horizontal mode. ...............................................11

Figure 6. Mean + SE nightly passage rates recorded in vertical mode.........................................12

Figure 7. Mean + SE hourly passage rates recorded in horizontal mode. ....................................13Figure 8. Mean + 1 SE hourly passage rates recorded in vertical mode.......................................14

Figure 9. Frequency histogram of targets by height class, sampling at 1.5-km. Height

class 1 represents altitudes 0-100om, class 2 represents altitudes 100-200

om, etc.

No targets were observed in classes 10-12, 14, or 15........................................................16

Figure 10. Mean + 1 SE nightly flight altitude sampling at 1.5-km range. ..................................17

Figure 11. Mean + 1 SE hourly flight altitude sampling at 1.5-km range....................................18Figure 12. Recorded target altitude distributions..........................................................................19

Figure 13. Mean + 1 SE nightly target air speed. .........................................................................21

Figure 14. Diurnal avian mean use estimates for each survey point by season at the CapeVincent wind power project area. ......................................................................................29

Figure 15. Breeding bird survey point count locations for the Cape Vincent wind powerproject area.........................................................................................................................32

Figure 16. AnaBat survey locations for the Cape Vincent wind power project area....................36Figure 17. Waterfowl and winter raptor driving transects with species location recorded

for the project area. ............................................................................................................41


8/59



WEST, Inc. 1 November 28, 2007

INTRODUCTION AND BACKGROUND

BP Alternative Energy North America, Inc. (BPAE) is evaluating the feasibility of wind energy

development in Jefferson County, New York. The proposed project,Cape Vincent Wind PowerProject, is located south of the St. Lawrence River and north of Chaumont Bay, near the town of

Cape Vincent, New York (Figure 1). The city of Watertown is located approximately 12 milessoutheast of the project. The exact location and size of the development will be based on a

number of factors including power purchase agreement(s), electricity markets, transmission

constraints, permitting, and results of site surveys.

Through the early project evaluation process, BPAE contacted the New York State Department

of Environmental Conservation (NYSDEC) and U.S. Fish and Wildlife Service (USFWS) todetermine biological resources of concern for the project. Issues that were raised included

potential impacts from the project on avian and bat resources, in particular nocturnal migrant

birds and migrant raptors, migrant bats, and species of concern that may occupy the site. Inresponse to comments from the agencies, BPAE requested that Western EcoSystemsTechnology, Inc. (WEST) develop an avian and bat survey protocol for a one-year study that

would address the agency concerns and provide site-specific data for the resources of concern.

The principal goals of the study, initiated in April 2006, were to:

1)Provide baseline information on avian and bat resources and use of the study area thatwould be useful in evaluating potential impacts from the wind power development;

2)Provide baseline information on avian and bat migration over the proposed developmentarea that is useful in evaluating the relative risk of the proposed wind project;

3)Provide information on avian, bat, and sensitive species use of the study area that willhelp in designing a wind project that is less likely to expose species to risk of collisions

with turbines, and;

4)Provide recommendations for further monitoring studies and potential mitigationmeasures, if appropriate.

Specific objectives of the study were to: (1) describe and quantify nocturnal migration over the

proposed project area; (2) describe and quantify spring and fall (diurnal) raptor migration

through the proposed project; (3) describe and quantify breeding bird use in the proposed

development area (turbine locations); (4) describe and quantify migrant bat use over theproposed project; (5) identify resident bat species in the project area; (6) describe and quantify

waterfowl migration through the project area; (7) and identify the presence of any federal and

state-listed species that may occur within in the project area, as well as potential habitat for thesespecies. The protocol was developed based on input from NYSDEC and the USFWS, as well as

the expertise and experience of WEST implementing and conducting similar studies for wind

energy development throughout the U.S.


9/59




Figure 1. Proposed Cape Vincent wind power projectlocation.


10/59




STUDY AREA

The proposed project area is located within the Great Lakes Plain ecozone in northern New York

(Andrle and Carroll 1988). Elevation of the ecozone varies from about 100-500 feet. The

dominant vegetation type was historically northern hardwood forest: oaks, beech, sugar maple,white ash, and black cherry; but agricultural clearing has left the region approximately 20%

wooded (Andrle and Carroll 1988). Some of the overall project area is characterized by Alvarecosystems: grasslands, shrublands, woodlands, and sparsely vegetated rock barrens that develop

on flat limestone where soils are very shallow (Edinger et al. 2002).

The land within the project area is privately owned and the primary land use is agriculture and

dairy farming (Figure 2). Most of the project development area is in agricultural fields. There

are scattered farms and houses throughout the project and adjacent to the roads. Vegetation ofthe project is a mosaic of open grass/hay fields, cultivated agriculture, and scattered deciduous

tree wood lots. The deciduous forest type tends to be variable in size with some small woodlots

intermixed with agriculture fields and some larger blocks of forest, particularly in low-lyingareas and along stream corridors. Several inlets, creeks, and wetland forests occur within theproject area.

STUDY COMPONENTS

The one-year avian and bat preconstruction study consisted of nocturnal marine radar samplingduring the spring and fall migration periods; diurnal point count surveys from fixed point

locations conducive to observing raptors and other large birds; breeding bird survey point counts;

AnaBat sampling for migrating bats during the spring and fall; AnaBat sampling for resident bats

during the summer; and winter and early spring waterfowl and raptor surveys. The various studycomponents took into consideration the potential for federal and state-listed species occurrence

in the project area.


11/59


12/59




Nocturnal Marine Radar Survey

The overall purpose of the nocturnal marine radar survey is to characterize avian migration over

the project area and provide data that can be used to determine the relative magnitude ofnocturnal migration over the proposed development area when compared to other sites. The

primary objective of the radar study is to collect baseline information on flight direction, passagerates, and flight altitude of nocturnal migrants at a representative sampling location for the

proposed development area.

A single radar unit was used for the migration seasons defined as 15 August 15 October for the

fall and 15 April 15 1 June for the spring. The radar lab consists of an X-band marine radar,

transmitting at 9,410 MHz with power output of 12 kW, mounted on a vehicle. Similar radarlabs have been successfully used to monitor nocturnal avian migration and are described in

Cooper et al. (1991) and Harmata et al. (1999).

The fall sampling location was selected based on constraints of the radar (e.g., minimization ofground interference), property ownership, access, and comments from the NYSDEC and

USFWS (Figure 3). Based on comments from the NYSDEC and USFWS, the ideal radar

sampling point to allow characterization of avian/bat movement along the shoreline, as well asover inland areas, was restricted to those areas approximately 1.5 km from the shoreline. To

decrease ground clutter, the unit was positioned in a small hollow so that surrounding

topography reflected the lower portion of the main beam, producing a clear picture of skybeyond. Due to land management changes at the fall radar sampling location, the site was

inaccessible in the spring. A second sampling location was chosen with similar characteristics as

the fall site and also situated approximately 1.5 km from the shoreline (Figure 3).


13/59



WEST, Inc. 6

Figure 3. Radar sampling locations and raptor survey locations for the Cape Vincent


14/59



WEST, Inc. 7 Draft - November 1, 2007

Methods

The study period for radar sampling was 63 days during the fall season and 50 days during the

spring. Due to the constraints of marine radar, sampling during some nights was compromised

or cancelled due to rain, so the total number of sampled nights was less than the total studyperiod. Nocturnal radar sampling occurred from approximately sunset each night until sunrise

the following morning. Each night was broken down into 60-min sampling periods thatconsisted of:

1)one 5-min session to collect weather data and adjust the radar to surveillance (i.e.,horizontal) mode,

2)one 10-min short-range session (1.5 km range) with the radar in surveillance modecollecting information on migration traffic (passage) rates;

3)one 10-min short-range session (1.5 km range) with the radar in surveillance modecollecting information on flight direction and speed of targets, as well as general

location of migrants;4)one 5-min break to adjust radar to vertical mode;5)one 10-min short-range session (1.5 km range) in the vertical mode to collect

information on migration traffic (passage) rate;

6)one 10-min short-range session (1.5 km range) in the vertical mode to collectinformation on flight altitudes below 1500 m;

7)one 5-min short-range session (1.5 km range) in the vertical mode to collectinformation on the spatial distribution and altitudes of birds along an east-westtransect axis; and,

8)one 5-min long-range session (3.0 km range) in the vertical mode to collectinformation on flight altitudes below 3000 m.

The following weather data was collected at the beginning of each hour session: wind speed,

wind direction; cloud cover (%); approximate ceiling height (m); approximate visibility (m);

precipitation; barometric pressure; air temperature (oC). Noticeable changes in weather

conditions, if any, were recorded when the radar unit was adjusted to vertical mode.

The Furuno FAR2117BB radar used in this study has several controls which affect detection andtracking of targets. In order to detect and track small targets, the radar operated under the

shortest pulse length setting with the gain control turned up to near the highest setting. Initially,

the anti-clutter controls on the radar were turned down to the lowest settings. The anti-sea clutter

control was then slowly turned up to about the point where background noise cleared from the

screen enough to see small targets. The anti-rain clutter control was kept at the lowest setting.While in the vertical mode, to eliminate ground clutter around the radar generated from second

echoes of radar energy bouncing off the van and ground, a blind sector was set so that the radardid not transmit energy when the antennae was pointing towards the ground (from 90 o to 270o).

The radar trails function was generally set at 30 seconds so that targets could be tracked for long

enough to determine direction and speed. Target flight direction was determined by placing thecursor on a target echo within a trail and aligning the offset electronic bearing line (EBL) along

the line of target echoes pointing in the direction of travel. Speed was recorded as the distance a


15/59




target traveled in 5 seconds (two sweeps of the radar antennae). With the target trails turned on,

each sweep of the radar plots a new echo for any given target with each echo persisting on the

screen for a set amount of time (e.g., 30 seconds). Speed was determined with the offset variable

range marker (VRM) by placing the cursor on a target echo and measuring the distance between

that echo and the third echo in line (i.e., the distance traveled in 2 sweeps of the antennae or 5seconds). Target height was measured with an index line (a tangent on the VRM) on the monitor

relative to a horizontal line running through the radar point of origin.

All data were exported from Microsoft Access and imported into SAS V.8 for further data

processing, quality assurance, and analysis. Additional analyses were performed using MatlabV6.5. To determine passage rates in horizontal mode, the 2-dimensional area represented by the

radar image was treated as a 1-dimensional front perpendicular to the direction of migration,

with length equal to 3 km (the diameter of the surveyed area); all targets counted in the radarimage during the sampling period were treated as if they had crossed the front. Based on that

assumption, passage rate was calculated as number of targets per kilometer per hour.

Mean flight direction was estimated as ( )1tan y x = where ( )

1cos

n

iiy n

== ,

( )1sin

n

iix n

== , and iwas the flight direction for the ith observation (Batschelet, 1981).

Dispersion in the data was calculated as ( )1 2

2 2r x y= + such that 0 r1. If all observations

had exactly the same direction, r= 1; conversely, r= 0 would indicate uniform distribution ofdirections around the circle.

Mean flight altitude was not adjusted for unequal sampling intensity at different heights or

unequal detection probability as a function of distance from the radar unit.

Air speed of targets, Va, was calculated as ( )2 2 2 cosa g w g wV V V V V = + , where Vg = target

ground speed, Vw = wind speed, and was the difference between the target flight directionand wind direction. Hourly weather observations made at ground level were used for estimates

of wind speed and direction. Wind direction categorized by field observers as N, NE, E,

SE, etc.; were transformed to bearings (0, 45, 90, 135, etc.) for the calculation of .Targets with air speeds less than 6 m/s or greater than 35 m/s were judged not to be migrating

birds and were excluded from further analysis.

Results

Nocturnal radar surveys were conducted most nights during the 63-day period between August

15 and October 15, 2006 and the 50-day period between April 19 and June 8, 2007. During fall,

radar sampling was conducted most nights for a total of approximately 508 hours of radarsampling during the study period. Very wet weather in mid-April and again in late-May

compromised many survey nights during the spring study period. Radar sampling was

conducted for a total of approximately 300 hours during the spring study period.


16/59




200

400

600

800

30

210

60

240

90270

120

300

150

330

180

0

Flight Direction

Observed flight directions were towards the southwest in the fall and towards the northeast in the

spring (Figure 4). Fall mean and dispersion of flight direction were= 209.2 and r= 0.34

(n = 12378 targets). As an indication of the southerly direction of the migration, 71.8% ofobservations were between 90 and 270, while 34.5% of observations were between 135 and225. Spring mean and dispersion of flight direction were= 34.0 and r= 0.52 (n = 5003

targets). As an indication of the northerly direction of the migration, 77.6% of observations were

between 270 and 90, and 48.4% of observations were between 315 and 45.

Figure 4. Observed flight directions at Cape Vincent project area.

Passage Rates

Fall -The overall mean passage rate in the horizontal mode was 345.8 13.3 targets/km/hr(mean SE) (n = 506 sample periods) and in the vertical mode was 346.2 17.2 targets/km/hr(mean SE) (n = 503 sample periods). Mean nightly passage rate was highly variable in bothhorizontal mode (Figure 5) and vertical mode (Figure 6). The greatest nightly passage ratesoccurred in late September and early October. Mean hourly passage rates tended to be low early

in the evening, with rapid increases to maximum values just before midnight, followed by

progressively declining rates throughout the night (Figures 7 and 8).


17/59




Spring -The overall mean passage rate in the horizontal mode was 166.2 8.8 targets/km/hr(mean SE) (n = 310 sample periods) and in the vertical mode was 191 9.4 targets/km/hr(mean SE) (n = 308 sample periods). Mean nightly passage rate was highly variable in bothhorizontal mode (Figure 5) and vertical mode (Figure 6). The greatest nightly passage rates

occurred in early and mid May. Mean hourly passage rates tended to be low early in theevening, with rapid increases to maximum values just before midnight, followed byprogressively declining rates throughout the night with a second small increase early in the

morning (Figures 7 and 8).


18/59




08/14 08/24 09/03 09/13 09/23 10/03 10/13

0

500

1000

1500

DATE

NIGHTL

YPASSAGERATE(targets/km/hr)

Fall

Spring

Figure 5. Mean + 1 SE nightly passage rates in horizontal mode.


19/59




08/14 08/24 09/03 09/13 09/23 10/03 10/13

0

200

400

600

800

1000

1200

1400

1600

1800

2000

DATE

NIGHTLYPASSAGERATE(targets/km

/hr)

Fall

Spring

Figure 6. Mean + SE nightly passage rates recorded in vertical mode.


20/59




1800 2000 2200 0000 0200 0400 0600

0

100

200

300

400

500

600

TIME

HOURLY

PASSAGERATE(targets/km/h

r)

Fall

Spring

Figure 7. Mean + SE hourly passage rates recorded in horizontal mode.


21/59




1800 2000 2200 0000 0200 0400 0600

0

100

200

300

400

500

600

TIME

HOUR

LYPASSAGERATE(targets/km/hr)

Fall

Spring

Figure 8. Mean + 1 SE hourly passage rates recorded in vertical mode.


22/59




Flight Altitudes

Fall - For sampling at the 1.5-km range in vertical mode, mean flight altitude was 490.4 1.7 m(mean SE) (n = 30,749 targets) above radar level (arl)1. Approximately 7.7% of targets hadflight altitudes less than 125 m (the approximate zone of risk posed by modern turbines) at the

site. Most targets were observed at altitudes below 500 m (Figure 9). The highest percentage oftargets occurred between 201 and 300 m arl. Nightly mean flight altitudes were variable

throughout the study period and ranged from approximately 275 m to 685 m arl (Figure 10). In

contrast, hourly mean flight altitudes were relatively constant (typically in the 450500 m range)(Figure 11) and close to the overall mean flight altitude for the study period. For sampling

periods at the 3-km range in vertical mode, 3.1% of targets (558 of 18,059) had flight altitudesgreater than 1500 m. On all sampling nights the mean flight height was greater than the median

value and the middle 50% of all observations were greater than 125 m arl (Figure 12).

Spring - For sampling at the 1.5-km range in vertical mode, mean flight altitude was 441.3 2.5m (mean SE) (n = 16,151 targets) arl. Approximately14.0% of targets had flight altitudes less

than 125 m. The highest percentage of targets (19.2%) occurred between 101 and 200 m arl(Figure 9). Nightly mean flight altitudes were variable throughout the study period and rangedfrom approximately 170 m to 650 m arl (Figure 10). In contrast, hourly mean flight altitudes

were relatively constant (typically in the 440470 m range) (Figure 11)and close to the overall

mean flight altitude for the study period. For sampling periods at the 3-km range in verticalmode, 2.6% of targets (253 of 9061 targets) had flight altitudes greater than 1500 m. On all

sampling nights the mean flight height was greater than the median value and above 125 m arl;

however, on two nights the median value was below 125 m arl and on seven nights the middle

50% of all observations overlapped the zone of risk (Figure 12).

1 Target altitude was measured in relation to a horizontal line running through the point of origin for the radar and

thus termed above radar level. Height above ground level (agl) is highly variable depending on the topography

directly below any given target and not measurable with the radar.


23/59




1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

2

4

6

8

10

12

14

16

18

20

HEIGHT CLASS

PERCENTOFTARGETS

1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

5

10

15

HEIGHT CLASS

PERCENTOFTARGETS

Fall

Spring

Figure 9. Frequency histogram of targets by height class, sampling at 1.5-km. Height class 1

represents altitudes 0-100om, class 2 represents altitudes 100-200

om, etc.


24/59




08/14 08/24 09/03 09/13 09/23 10/03 10/13

0

50

100

150

200

250

300

350

400

DATE

FLIGHTALTITUDE(m)

04/26 05/01 05/06 05/11 05/16 05/21 05/26 05/31 06/05 06/10

0

100

200

300

400

500

600

700

DATE

FLIGHTALTITUDE(m)

Fall

Spring

Figure 10. Mean + 1 SE nightly flight altitude sampling at 1.5-km range.


25/59




1800 2000 2200 0000 0200 0400 0600

0

50

100

150

200

250

300

TIME

FLIGHTALTITUDE(m)

0000 0200 0400 2000 2200

0

50

100

150

200

250

300

350

400

450

500

TIME

FLIGHTALTITUDE(m)

Fall

Spring

Figure 11. Mean + 1 SE hourly flight altitude sampling at 1.5-km range.


26/59




Fall

Spring

Figure 12. Recorded target altitude distributions2.

2 The boxes within the chart represent the 1st and 3rd quartile (50%) of the nightly observations, the horizontal lines

within boxes represent nightly median value of flight heights, and solid circles represent the nightly mean flight

height.


27/59




Target Speed

Fall - Air speed of targets was calculated by adjusting for wind speed and direction (see Methods

above). Of 12,190 targets, approximately 1% (120 targets) were moving very slow (< 6 m/s) andone target was moving at high speed (> 35m/s). After excluding very slow and very fast targets,

overall mean target air speed was 12.95 0.03 m/s (mean SE) (n = 12069 targets). Nightlymean target air speed varied from approximately 10 to 17 m/s (Figure 13). Because thepercentage of targets moving slowly was so small, no further adjustment to the data set was

warranted.

Spring - Of 5,003 targets, approximately 1% (56 targets) was excluded because they were

moving very slow (< 6 m/s) or due to high speed (> 35m/s) and 47 targets were excluded due to

missing wind speed and/or direction to allow for air speed adjustments. After excluding very

slow and very fast targets, overall mean target air speed was 13.65 0.06 m/s (mean SE)(n = 4900 targets). Nightly mean target air speed varied from approximately 11 to 18 m/s(Figure 13). Because the percentage of targets moving slowly was so small, no further

adjustment to the data set was warranted.


28/59




08/14 08/24 09/03 09/13 09/23 10/03 10/13

0

2

4

6

8

10

12

14

16

18

DATE

AIRSPEED(m/s)

Fall

Spring

Figure 13. Mean + 1 SE nightly target air speed.


29/59


30/59




more inland in eastern portions of the state during fall migration. According to spring count data

from the Derby Hill Bird Observatory (Mexico, New York) approximately 50 miles south of

Cape Vincent along Lake Ontario, peak numbers of sharp-shinned hawks migrate through thearea during April, with large pulses of broad-winged hawks during the last two weeks of April.

Fall migration counts from Franklin Mountain in Oneonta, New York (150 miles southeast ofCape Vincent) report peak periods for migrant broad-winged and sharp-shinned hawks during

September and October, respectively. Concern for migrant golden eagles potentially using theCape Vincent project area was expressed during talks with the NYSDEC. Golden eagles are

earlier and later migrants with peaks reported from the end of March through April during spring

migration and the end of October through November during fall migration. Spring raptorsurveys at the Cape Vincent project area began later in the 2006 season (April 14, 2006) and

likely did not capture early raptor migrants, such as golden eagles; however, spring surveys were

conducted again in 2007 and began at an earlier date, March 21, and ran until May 1. In fall,surveys were conducted from September 23 November 11.

ResultsDuring the spring 2006 season, each point was surveyed 4 times, for a total of 12 surveys. A

total of 777 individual birds were recorded; 79 raptors of 10 species were observed (Table 1).

During the fall season, each fixed point was surveyed 10 times during the survey window, for atotal of 30 surveys. A total of 3,050 individual birds were recorded during the surveys; 165

individual raptors of 10 species were observed. During the spring 2007 season, each point was

surveyed 7 times, for a total of 21 surveys. A total of 1,851 individual birds were recordedduring the surveys; 205 individual raptors of 9 species were observed. (Table 1)

Canada goose was the most commonly seen bird during spring and fall surveys. During both

spring migration surveys (2006 and 2007), turkey vulture was the mostly commonly recordedraptor species (n = 29, freq = 66.7% and n = 111, freq = 94.4%, respectively). American kestrel

(n = 13, freq = 41.7%) followed turkey vulture in numbers during the spring 2006 surveys,

whereas northern harrier (n = 37, freq = 88.9%) followed turkey vulture in the 2007 springsurveys. In the fall, northern harrier was the most commonly recorded raptor species (n = 69,

freq = 76.7%), followed by turkey vulture (n = 50, freq = 33.3%). Other raptor species seen

included: broad-winged hawk, red-tailed hawk, rough-legged hawk, sharp-shinned hawk, osprey,peregrine falcon, and Coopers hawk.

Exposure indices were calculated as the mean use estimates (number of birds/60-minute survey)

multiplied by the proportion of birds observed flying and the proportion of birds flying within

the zone of risk (defined as the approximate rotor-swept area). During the migratory seasons,gull species had the highest exposure index due to high numbers of individuals occurring in the

project area (Table 2). For raptors, turkey vulture had the highest exposure index also dueprimarily to the higher use estimates.

Avian and raptor use varied among survey stations (Figure 14). Avian use was highest at Station1 during the fall 2006 season; however Station 2 was the highest during both spring seasons.

Large flocks of Canada geese and other duck species recorded at this survey point contributed to


31/59




this high use estimate. Station 2 is located on the western edge of the project area and closest to

Lake Ontario. High numbers of Canada geese and gull species accounted for higher avian use at

this survey station. Raptor use was generally similar between seasons and survey points. Station2 had higher use in the spring seasons but the differences were not significant.


32/59



WEST, Inc. 25

Table 1. Raptors and other large bird species observed during spring and fall diurnal raptor migration sur

power project area.

Spring 2006 Fall 2006

Species/Group# ind

# groups

mean

use3

% freq

4# ind

# groups

mean

use% freq #

Waterbirds 221 22 34 18

Bonaparte's gull 0 0 0.00 0.00 0 0 0.00 0.00

Caspian tern 0 0 0.00 0.00 0 0 0.00 0.00

Common loon 0 0 0.00 0.00 1 1 0.03 3.33

Common tern 0 0 0.00 0.00 0 0 0.00 0.00

Double-crested cormorant 0 0 0.00 0.00 0 0 0.00 0.00

Great blue heron 8 7 0.67 50.00 3 3 0.10 10.00 2

Herring gull 6 2 0.50 16.67 6 2 0.20 6.67

Ring-billed gull 57 6 4.75 33.33 8 7 0.27 20.00 2

Unidentified gull 150 7 12.50 41.67 16 5 0.50 13.33

Waterfowl 457 25 2677 92

Canada goose 411 19 34.25 75.00 2337 64 77.90 50.00 13

Bufflehead 0 0 0.00 0.00 9 1 0.30 3.33

Common merganser 0 0 0.00 0.00 9 2 0.30 6.67

Gadwall 0 0 0.00 0.00 14 1 0.47 3.33

Green-winged teal 0 0 0.00 0.00 2 1 0.07 3.33

Hooded merganser 0 0 0.00 0.00 26 3 0.87 6.67

Mallard 41 5 3.42 25.00 91 16 3.03 40.00

Ring-necked duck 0 0 0.00 0.00 1 1 0.03 3.33

Tundra swan 0 0 0.00 0.00 20 1 0.67 3.33

Unidentified duck 5 1 0.42 8.33 168 2 5.60 6.67 Raptors 79 58 165 129

Accipiters

3

Mean use = number observed within 800 m of survey point per 60-min survey4

Frequency of occurrence = percent of surveys in which species was observed


33/59



WEST, Inc. 26

Spring 2006 Fall 2006

Species/Group# ind

# groups

mean

use3

% freq4

# ind

# groups mean

use% freq #

Coopers hawkSC 0 0 0.00 0.00 3 3 0.10 10.00

Northern goshawk 0 0 0.00 0.00 0 0 0.00 0.00 Sharp-shinned hawkSC 3 3 0.25 25.00 1 1 0.03 3.33

Buteos

Broad-winged hawk 8 6 0.67 33.33 0 0 0.00 0.00

Red-tailed hawk 11 10 0.92 50.00 29 23 0.97 33.33

Rough-legged hawk 2 2 0.17 16.67 2 1 0.07 3.33

Unidentified buteo 1 1 0.08 8.33 2 1 0.07 3.33

Falcons

American kestrel 13 5 1.08 41.67 5 5 0.17 16.67

Peregrine falcon 0 0 0.00 0.00 2 2 0.07 6.67

Other Raptors

Northern harrierST 7 7 0.58 33.33 69 63 2.30 76.67

OspreySC

1 1 0.08 8.33 0 0 0.00 0.00

Turkey vulture 29 19 2.42 66.67 50 28 1.67 30.00 1

Unidentified raptor 4 4 0.33 25.00 2 2 0.07 6.67

Other Birds 20 8 170 65

American crow 20 8 1.67 41.67 146 56 4.83 70.00

Common raven 0 0 0.00 0.00 1 1 0.03 3.33

European starling 0 0 0.00 0.00 0 0 0.00 0.00 1

Killdeer 0 0 0.00 0.00 4 2 0.13 6.67

Ring-necked pheasant 0 0 0.00 0.00 8 6 0.27 16.67

Rose-breasted grosbeak 0 0 0.00 0.00 0 0 0.00 0.00 Wild turkey 0 0 0.00 0.00 15 2 0.50 6.67

Total 777 113 3050 306

ST = State listed threatened; SC = State listed species of special concern


34/59



WEST, Inc. 27

Table 2. Flight height characteristics and exposure indices by species observed during diurnal raptor migra

wind power project area.

Relation to rotor-swept area

SpeciesMean Use

% birds

flying% below % within %

Waterbirds 97.47 28.10 29.32 Bonaparte's gull 0.04 0.00 NA NA Caspian tern 0.02 100.00 0.00 100.00

Common loon 0.01 0.00 NA NA

Common tern 0.01 100.00 0.00 100.00 Double-crested cormorant 0.01 100.00 0.00 0.00

Great blue heron 0.44 97.30 27.78 61.11

Herring gull 0.15 100.00 41.67 58.33

Ring-billed gull 1.06 98.84 81.18 17.65 Unidentified gull 2.10 100.00 42.69 56.73

Waterfowl 98.09 50.97 46.75 Canada goose 51.00 98.64 28.78 29.87 Bufflehead 0.15 75.00 100.00 0.00

Common merganser 0.11 100.00 22.22 0.00

Gadwall 0.17 100.00 0.00 100.00 Green-winged teal 0.02 100.00 0.00 100.00

Hooded merganser 0.38 54.84 100.00 0.00

Mallard 2.11 84.80 36.55 48.28 Ring-necked duck 0.16 0.00 NA NA

Tundra swan 0.25 100.00 0.00 0.00

Unidentified duck 2.19 97.74 0.00 2.89

Raptors 95.05 30.63 41.46 Accipiters 100.00 10.00 50.00 Cooper's hawk 0.06 100.00 20.00 60.00

5

Defined as the area between approximately 25 and 125 m above ground level6

Exposure index = (mean use) * (% individuals flying) * (% flying within rotor-swept area)


35/59



WEST, Inc. 28

Relation to rotor-swept area

SpeciesMean Use

% birds

flying% below % within %

Northern goshawk 0.01 100.00 0.00 0.00

Sharp-shinned hawk 0.05 100.00 0.00 50.00 Buteo 88.97 21.49 42.98 Broad-winged hawk 0.10 100.00 0.00 62.50 Red-tailed hawk 0.94 90.00 18.06 50.00

Rough-legged hawk 0.47 81.58 41.94 35.48

Unidentified buteo 0.06 100.00 0.00 0.00

Falcon 76.92 93.33 6.67

American kestrel 0.46 75.68 100.00 0.00

Peregrine falcon 0.02 100.00 0.00 100.00

Owls

Short-eared owl 0.02 100.00 100.00 0.00

Other raptorsNorthern harrier 1.46 100.00 66.67 24.17 Osprey 0.02 100.00 0.00 50.00

Turkey vulture 2.20 99.47 4.76 57.14

Unidentified raptor 0.07 100.00 16.67 33.33

Other BirdsAmerican crow 3.56 87.20 51.59 38.10

Common raven 0.07 100.00 66.67 33.33 European starling 1.36 68.18 100.00 0.00

Killdeer 0.11 100.00 44.44 55.56

Ring-necked pheasant 0.26 19.05 100.00 0.00

Rose-breasted grosbeak 0.09 100.00 0.00 100.00 Wild turkey 0.52 4.76 100.00 0.00


36/59



WEST, Inc. 29

Figure 14. Diurnal avian mean use estimates for each survey point by season at the Cape Vincent w

All Birds for Spring 2006

0

10

20

30

40

50

60

70

80

90

1 2 3

Station

Meanuse

Raptors for Spring 200

0

3

6

9

12

15

1 2

Station

Meanuse

All Birds for Fall 2006

0

50

100

150

200

250

300

1 2 3

Station

Meanuse

Raptors for Fall 200

0

3

6

9

12

15

1 2

Station

Meanuse


37/59


38/59




Breeding Bird Survey

The objective of the breeding bird surveys was to estimate the spatial and temporal use of theproposed development area by breeding resident birds. The emphasis of the surveys was

locating and counting breeding resident birds within the area proposed for development. Thesurveys were conducted based on the regional timing recommended for USGS BBS in central

New York (USGS 2001).

Methods

Twenty survey points were established within the project area. The survey points were selected

to cover as much of the proposed development area and habitat types as possible. Each survey

station was marked on a map and GPS coordinates were recorded for mapping (Figure 15). Thehabitat at each survey point was described to examine the applicability of the site to represent

other areas within the proposed development area.

U.S. Geological Survey Breeding Bird Survey (USGS 2001) methods were used for the surveys.Each survey plot was a variable circular plot centered on the observation point. All birds

observed were recorded; however, the survey effort was concentrated within an approximate 400

m (0.25 mi) radius circle centered on the observation point. All points were surveyed twiceduring the recommended survey period (June - July) and seven days were skipped between the

surveys to spread the effort over the breeding season.

Survey periods at each point were 3 minutes long, similar to the BBS method. The date; start

and end time of the observation period; and weather information such as temperature, wind

speed, wind direction, and cloud cover were recorded for each survey. Species or best possible

identification, number of individuals of each species, how observed (visual or auditory), andbehavior (flying, perching, singing, etc.) were recorded for each observation during the 3-minute

count at each survey point.


39/59




Figure 16. Breeding bird survey point count locations for the Cape Vincent wind power project

area.


40/59




Results

Point count surveys were conducted on June 29 and July 6, 2006. Each point was surveyedtwice, for a total of 40 survey periods. A total of 812 individual birds were observed in 462

groups (Table 3). Sixty-three species were observed during the surveys. Red-winged blackbird,bobolink, and song sparrow were the most common passerines observed based on mean use

estimates (number observed within 400 m per 3-minute survey). The majority of the speciesrecorded during breeding bird surveys are species commonly associated with agriculture,

grasslands, and/or edge habitat. Several species of interest were recorded during the breeding

bird surveys including northern harrier and Henslows sparrow, two New York state threatenedspecies; horned lark, grasshopper sparrow, and vesper sparrow, three New York state species of

concern; and bobolink and wood thrush, two species on the USFWS 2002 Birds of Conservation

Concern list for the Lower Great Lakes/St. Lawrence Plain region.

Table 3. Avian species observed during breeding bird surveys within the Cape Vincent windpower project area.

Species/Group # of individuals # of groups Mean UseWaterbirds 46 10 1.15

Double-crested cormorant 2 1 0.05

Great blue heron 6 5 0.15

Ring-billed gull 38 4 0.95

Waterfowl 55 3 1.375

Canada goose 12 1 0.3

Mallard 8 1 0.2

Unidentified duck 35 1 0.875

Shorebirds 22 8 0.55Killdeer 7 7 0.175

Unidentified shorebird 15 1 0.375

Raptors/Vultures 41 30 1.025American kestrel 9 8 0.225

Northern harrierST 8 6 0.2

Red-tailed hawk 6 5 0.15

Turkey vulture 18 11 0.45

Passerines 612 394 15.095

American crow 41 16 1.025

American goldfinch 27 17 0.675

American redstart 2 2 0.05

American robin 31 26 0.775Baltimore oriole 1 1 0.025

Barn swallow 19 6 0.475

Black-and-white warbler 3 2 0.075

Black-capped chickadee 8 4 0.02

Blue jay 6 6 0.15

BobolinkBCC 65 38 1.625

Brown-headed cowbird 8 5 0.2


41/59




Species/Group # of individuals # of groups Mean UseCarolina wren 1 1 0.025

Cedar waxwing 27 8 0.675

Chestnut-sided warbler 10 10 0.25

Chipping sparrow 2 2 0.05Common grackle 22 9 0.55

Common yellowthroat 21 16 0.525

Eastern bluebird 1 1 0.025

Eastern kingbird 20 13 0.5

Eastern meadowlark 26 24 0.65

Eastern towhee 19 16 0.475

Eastern tufted titmouse 2 1 0.05

Empidonax sp. 1 1 0.025

European starling 13 4 0.325

Field sparrow 2 2 0.05

Grasshopper sparrowSC

2 2 0.05

Gray catbird 6 6 0.15

Henslows sparrowST, BCC 1 1 0.025

House finch 1 1 0.025

Horned larkSC 3 2 0.05

Indigo bunting 1 1 0.025

Northern cardinal 3 2 0.075

Ovenbird 4 3 0.1

Red-eyed vireo 6 6 0.15

Red-winged blackbird 75 38 1.875

Savannah sparrow 19 15 0.475

Scarlet tanager 2 2 0.05

Song sparrow 55 44 1.375

Tree swallow 4 2 0.1

Unidentified passerine 7 1 0.175

Unidentified sparrow 5 3 0.125

Vesper sparrowSC

1 1 0.025

Wood thrushBCC

3 3 0.075

Yellow warbler 38 32 0.95

Upland Gamebirds 3 3 0.075Ring-necked pheasant 2 2 0.05

Ruffed grouse 1 1 0.025

Doves 24 6 0.6

Mourning dove 5 3 0.125

Rock pigeon 19 3 0.475

Other Birds 6 6 0.15

Downy woodpecker 1 1 0.025Northern flicker 3 3 0.075

Red-bellied woodpecker 1 1 0.025

Unidentified woodpecker 1 1 0.025

All Birds 812 462 20.02

ST = State listed threatened; SC = State listed species of special concern; BBC = Birds of Conservation Concern


42/59




Nocturnal AnaBat SurveysThe objective of the nocturnal AnaBat surveys was to record the relative abundance of echo-

locating bats flying through the sampling area during summer breeding season and the spring andfall migration seasons.

Methods

Bat activity at the project area was recorded using an AnaBat II ultrasonic bat detector attachedto a zero-crossing analysis interface module (ZCAIM) which houses a compact flash memory

card for temporary download of ultrasonic activity files. To sample continuously on remote

mode (automatic data collection), the detector and ZCAIM were powered by an external 12Vbattery. Each AnaBat unit (detector, ZCAIM, and 12V battery) was enclosed inside a plastic box

or dry bag with the detector microphone positioned against a PVC tube protruding from the

box/bag. This design prevented water from damaging the AnaBat units without compromisingthe ability of the unit to detect ultrasonic noise in the environment. To minimize variationamong AnaBats, sensitivity settings were calibrated for each unit prior to data collection. Most

AnaBat units were set at or near setting 7 on the sensitivity dial. Each AnaBat unit was

positioned so that the microphone faced the same cardinal direction, east, for each samplingperiod. Calls were recorded from approximately sunset to sunrise (1900 0700). AnaBat units

were removed from the field approximately once per week to download files, recharge batteries,

and troubleshoot technical problems. Data gathered from the passive AnaBat units at the mettower were used to calculate bat activity (designated as number of calls/detector-night) present at

the site during the sampling periods. Nights that experienced any number of technical

difficulties were not included in the final analyses.

During the spring sampling season (April 13 June 2), three AnaBat sampling locations were

established (Figure 16). One unit was placed in the open grassy field at the project met tower

and two other units were deployed near wooded riparian areas within the project to increaselikelihood of detecting additional species. One of these riparian units, centrally located in the

project area, was stolen in late spring and never recovered. The remaining two sampling

locations (Met tower and Riparian 1) were maintained through the summer sampling season(June 28 August 8). During fall (August 13 October 9), two pulley systems were attached to

the met tower guy wires which allowed AnaBat units to be deployed at three different levels:

ground level (1 m above ground), approximately 25m high (half way up the met tower), and

approximately 50m high (near the top of the met tower).


43/59



WEST, Inc. 36

Figure 16. AnaBat survey locations for the Cape Vincent wind power project


44/59




In addition to the stationary passive units, a roaming or mobile AnaBat unit was deployed

during the summer to assess resident/breeding bat species present within the project area (Figure

16). Roaming sampling was conducted using a handheld AnaBat unit for 9 nights (3 samplingperiods of 3 consecutive nights each) at habitats likely to have high numbers of resident bats. To

select locations for active sampling, reconnaissance visits were made to the project area duringthe day time to select sampling locations based on the presence of travel corridors (trails and

roads), linear landscape features (forest edges), and access to water; habitat features known to beimportant for bats. Active sampling was conducted from sunset until 4 hours after sunset

(approximately 2100 0100).

Analysis of bat calls was conducted using Analook software (DOS version). Analook displays

ultrasonic activity in a format similar to a sonogram used for analysis of bird vocalizations (e.g.,

frequency versus time). Species identification was aided by the Preliminary Key to theQualitative Identification of Calls within the AnaBat System (Amelon 2005, unpublished data)

where characteristics such as slope, frequency, minimum frequency, consistency of minimum

frequency, and shape of pulse assist in the identification of bat vocalizations. Due to similarityof call characteristics, two species (big brown and silver-haired bat) were lumped into onespecies category. All Myotis-like calls were identified to genus only and submitted to

NYSDEC-recommended biologist, Eric Britzke, for identification to species. To obtain species

identifications, an ID filter (Britzke and Murray 2001) was loaded into Analook to determinecalls sequences of sufficient quality and length for possible species identification. Once

separated, echolocation calls of sufficient quality and length were categorized using quantitative

techniques (Britzke 2003). Quantitative analyses are conducted by a cross-validatedclassification model based on 10 extracted call parameters [duration (Dur), maximum frequency

(Fmax), minimum frequency (Fmin), mean frequency (Fmean), duration to the knee (Tk),

frequency of the knee (Fk), duration of the body (Tc), frequency of the body (Fc), initial slope

(S1), and slope of the body (Sc)] collected from 1,846 sequences (35,979 calls) of 12 easternU.S. bat species (Britzke 2003). Average accuracy rates for species identification using this

statistical method ranges from 56.9% (eastern red bat) to 98.5 % (gray bat), with accuracy rates

for Indiana bat ranging from 81.4% to 88.6%.

Results

The total number of calls and number of calls per night, recorded by each AnaBat unit at the met

tower varied by season (Table 4). Spring sampling began on April 13, 2006 and recorded

continuously until June 2, 2006. The AnaBat unit detected 241 bat calls total (4.92 calls/night)

during the 49 days of spring sampling. Summer sampling occurred at the met tower on 15 nights

and recorded a total of 431 calls (28.73 calls/night). During fall, sampling occurred at 3 differentheights at the met tower. The AnaBat unit positioned at ground level recorded the highest

number of bat vocalizations per night (9.90 calls/night). Despite a similar number of samplingdays, the low position AnaBat unit recorded significantly more bat calls/night than either the

mid- or high-level units.


45/59




Table 4. Number of sampling days, total number of calls recorded, and calls/night recorded by

each AnaBat unit at the met tower for spring, summer, and fall sampling periods.

Season

Location

# of sampling

days used in

analysis

Total # of

calls

# calls/nightSpring Met tower low

49 241 4.92

Summer Met tower low 15 431 28.73

Fall Met tower low

mid

high

48

48

51

475

205

33

9.90

4.270.65

At least four species of bats were recorded at the met tower location (Table 5). Due to similarityof call characteristics, two species (big brown and silver-haired bat) were lumped into onespecies category. As is typical with AnaBat sampling, the majority of vocalizations were unable

to be identified due to the few number of pulses per call (


46/59




Summer sampling with the mobile AnaBat unit occurred on nine nights and recorded 316 bat

calls. The objective of the mobile sampling was to identify, to the extent possible, species and

relative abundance of each species using the Cape Vincent project area. No additional specieswere recorded during the roaming surveys that were not recorded at the met tower station. As

with the fixed station sampling, the majority of the calls could not be identified to species. Thehighest number of recorded calls was of big brown bat (Table 6); however, >50% of those calls

occurred on one night at one location and may have been from only one or a few individualsecho-locating repeatedly near the AnaBat microphone.

Table 6. Number of detections by species during summer roaming AnaBat sampling.Species Date Sampled

Common

Name

Scientific

Name

6/284 hrs

6/294 hrs

6/304 hrs

7/244 hrs

7/254 hrs

7/264 hrs

8/064 hrs

8/074 hrs

8/084 hrs

Big brown bat

Eptescus fuscus 8 3 0 33 7 0 8 2 0

Eastern red bat

Lasiurus borealis 0 1 0 0 0 7 2 0 2

Hoary bat

Lasiurus cinereus 0 0 0 3 0 0 0 0 0

Myotis species

Myotis spp. 0 0 0 0 0 2 0 0 0

No Species ID 42 17 15 48 8 10 41 53 3

Total Detections/night 50 22 15 84 15 19 51 55 5

Following the qualitative screening, 203 call files with characteristics resembling Myotis species

were submitted to Eric Britzke for further analysis. Of those files, 83 calls (40.9%) did not

contain sufficient enough information to be processed quantitatively. The remaining calls wereanalyzed quantitatively on a nightly basis by site (Britzke 2003). Calls meeting the quantitative

criteria for the following species were identified: eastern red bat (36 calls), little brown bat (44

calls), Indiana bat (25 calls), and northern myotis (15 calls).

Winter Waterfowl and Raptor Surveys

The objective of the waterfowl and winter raptor surveys was to estimate spatial and temporal

use of the site by migrant and wintering waterfowl and raptor species. During initial projectscoping, the agencies raised concerns over the potential for the proposed wind project to impact

wintering waterfowl and raptors.

Methods

Driving transect surveys were conducted along most roads through the proposed project area that

allowed nearly complete coverage of the project area (Figure 17). Surveys consisted of driving

transects to locate and count winter waterfowl in the project area. In addition, nine 30-minute

point count surveys were conducted at each of the fixed point count stations that were used


47/59


48/59



WEST, Inc. 41

Figure 17. Waterfowl and winter raptor driving transects with species location recorded for


49/59


50/59




DISCUSSION

Nocturnal Marine Radar Survey

The nocturnal radar study was designed to collect data that could be used to characterize

nocturnal migration over the site and also be used in a larger statewide comparison of resultsfrom numerous sites (M. Woythal, NYSDEC, pers. comm.). In the analysis, the radar data were

not corrected for differences in detectability with distance from the radar unit or due to ground

clutter on the radar screen. Also, the 2-dimensional area represented by the radar image wastreated as a 1-dimensional 3-km front perpendicular to the direction of migration, and all

targets counted in the radar image during the sampling period were treated as if they had crossed

the front. Thus, passage rate estimates should be considered a sample or index of the actualnumber of targets passing through the area.

Measurements from radar studies potentially are highly variable due to a number of factorsincluding observer bias and the radar settings affecting target detection. To minimize thesebiases, efforts were made to standardize data collection and radar settings as much as possible.

For example, the radar was operated under the shortest pulse length setting with the gain control

turned up to near the highest setting. While short wave-length and high gain insure detection ofsmall targets, these settings also have the effect of producing atmospheric or background noise

on the screen which consequently can obscure small targets. To clean up the screen the anti-

sea clutter [which minimizes clutter and noise close to the radar] was slowly turned up to thepoint where background noise was dispersed and limited primarily to the outer edge of the

screen. The anti-rain clutter [which reduces interference from small targets throughout the

survey area (e.g., rain drops)] was kept at the lowest setting so that no small targets would be

eliminated. These settings insure that small targets such as individual passerines can be detectedby the radar. Also during sampling, specific functions or capabilities of the radar were used to

determine data values to minimize observer bias. For example, the electronic bearing line and

variable range marker used in offset mode allowed the compass bearing of a target trail and thespeed at which the target was moving to be measured by the radar as opposed to estimated by the

observer or measured with a hand held scale.

Results from the nocturnal radar study conducted at the Cape Vincent project area were similar

to other radar studies in New York and the eastern U.S. (Table 9). Mean fall flight direction forthe Cape Vincent project area was 209 and for the spring was 34, slightly more southwesterly

and northeasterly than most other New York studies but within the range of directions reported at

other New York sites. Mean passage rate for fall 2006 was higher (346 t/km/hr) than the averagefor NY and the eastern U.S. (259 t/km/hr); however, it fell within the overall range of passage

rates reported at other New York sites. Conversely, spring passage rate was on the lower end of

the range of other studies. Mean flight height of targets was approximately 490 m in the fall and

441 m in the spring, which is similar to other studies in NY and near the means for all reportedstudies in the eastern U.S. (Table 9). The percent of targets (~8% fall and ~14% spring) which

flew through the zone of risk, defined as the air space below 125 m, were also very near the

mean for all other studies where flight height was recorded with vertical mode radar.


51/59


52/59




conducted in New York, while results have been variable, the highest fall passage rates have

been recorded at interior sites. For spring migration results again were variable with the highest

passages rates coming from a coastal site as well as two interior sites (see Table 9). The resultsfrom the Cape Vincent study do not appear to support the hypothesis that nocturnal migrants

concentrate along the shoreline.

The passage rates in the study area may have been influenced locally by the close proximity ofthe radar unit to the shoreline (


53/59




Based on the topography in the Cape Vincent peninsula area and Jefferson County there is little

to concentrate migrant raptors moving north and the study results appear to indicate that raptor

migration is more dispersed in the project region.

Table 10. Number of raptors observed per surveyor hour in the project area and at sixestablished New York spring/fall hawk watch sites.

Spring 2006 Cape VincentWind Project

Ripley Hawk Hamburg BraddockBay

DerbyHill

4/14/06 6.7 31.4 83.8 no survey 21.5

4/21/06 10.3 35.9 17.9 no survey 353.1

5/02/06 3.3 17.3 0.8 no survey 6.05/12/06 6.0 5.6 5.2 no survey 44.8

Average 6.5 22.5 26.9 -- 106.3

Spring 2007 Cape VincentWind Project

Ripley Hawk Hamburg BraddockBay

DerbyHill

3/21/07 3.0 23.8 7.1 25.2 77.9

3/31/07 18.0 27.9 123.5 53.5 74.1

4/11/07 11.3 31.0 19.2 38.4 71.74/14/07 1.0 31.4 83.8 95.1 81.1

4/17/07 6.0 2.0 1.09 no survey no survey

4/20/07 8.7 44.2 26.2 101.6 43.0

4/22/07 11.0 96.0 82.1 156.1 111.55/01/07 12.3 39.3 0.0 no survey 66.4

Average 9.8 37.0 42.9 78.3 75.1

Fall 2006 Cape Vincent Franklin Mt. Mohonk Preserve Mount Peter

9/23/06 3 1 no survey 1

9/30/06 7 3 2 5

10/07/06 12 10 no survey 3

10/13/06 9 3 11 7

10/20/06 2 no survey no survey no survey

10/27/06 9 20 11 5

10/30/06 5 15 16 10

11/05/06 4 1 no survey 1

11/07/06 2 0 no survey 2

11/11/06 2 2 no survey no survey

Average 5.5 9 10 3.4Daily count data acquired from HMANA 2006.

There are no fall hawk watch sites along the lake shoreline in central New York. The nearest fallsite, Kestrel Haven located in south central New York, was lower than the Cape Vincent project

area in terms of raptors counted per surveyor hour; however, count data for this site is only


54/59




available for 2005 so a direct comparison of actual survey days could not be made. Fall hawk

watch sites further south and east, such as Franklin Mountain, record similar numbers of migrant

raptors which are likely taking advantage of ridgelines of the western Appalachian Mountains;however, timing is different among sites. Higher numbers of raptors per surveyor hour were

seen earlier in the fall season at the Cape Vincent project area than at more southern sites. Thismay be a reflection of the more northern latitude of the study area or summer residents, such as

red-tailed hawk, turkey vulture, and northern harrier, still in the area.

Exposure indices are a common method for estimating risk to individual species from wind

turbines. During both migratory seasons, non-raptor species had the highest exposure index dueto high use of the area by waterfowl and waterbirds, such as Canada goose and gull species

(Table 2). At the Cape Vincent project area, raptors in general did not have high exposure

indices due to either low numbers recorded or flight heights outside of the zone of risk. Turkeyvulture had the highest exposure index; they were commonly observed and were most often

observed flying in the zone of risk. While these species have been recorded as fatalities at other

monitored wind plants, the number of fatalities are relatively small (see Erickson et al. 2001,2002). Red-tailed hawk was seen less frequently but was often seen flying in the zone of risk. Incontrast, northern harrier were often recorded, particularly during fall migration, but rarely

observed flying into the zone of risk and is rarely recorded as fatalities at other monitored wind

facilities (see Erickson et al. 2001, 2002).

Breeding Bird SurveyThe results of the breeding bird surveys were typical of agricultural settings in central New York.

Frequently recorded species included bobolink, red-winged blackbird, and song sparrow. A few

woodland species, such as wood thrush and ovenbird, were observed in small wooded areas and

wetlands scattered throughout the project area. Several species of gulls and waterfowl are alsopresent in the area due to the proximity to the shoreline. The closest breeding bird survey

(Watertown; Sauer 2005) reported similar species occurrences and abundances. Five species

listed by the NYSDEC were observed within the Cape Vincent project area: northern harrier,Henslows sparrow, horned lark, grasshopper sparrow, and vesper sparrow. Northern harrier and

Henslows sparrow are listed as state threatened species. The remaining three species are listed

as Special Concern species for New York (NYSDEC 2003). Bobolink, Henslows sparrow, andwood thrush are included on the 2002 Birds of Conservation Concern list for Lower Great

Lakes/St. Lawrence Plain region (USFWS 2002) in which the Cape Vincent project area occurs.

Based on the breeding bird survey data collected in 2006, the Cape Vincent project area does not

appear to have any large or unusual populations of breeding resident birds. Mortality resultsfrom two other eastern wind plants studied indicate that turbines on eastern mountain ridgelines

result in between 4 and 8 bird fatalities per turbine per year (see Kerns and Kerlinger 2004 andNicholson 2002, 2003). In both these studies it was estimated that approximately two-thirds of

the avian fatalities were migrants. Provided impacts at the Cape Vincent project area are similar,

it is not expected that breeding resident birds are at great risk from the wind project. Due to thediversity of birds recorded in the mixed farmland habitat, impacts are expected to be spread over


55/59




several commonly observed species (see Table 3). Currently, turbine layout is unknown;

therefore, potential impacts to breeding habitat of sensitive species are difficult to predict.

Nocturnal AnaBat Surveys

To date monitoring studies of wind projects have shown a few common trends in bat mortality.

Risk to bats from turbines appears to be unequal across species and seasons where increasedmortality occurs during the post-breeding or fall migration season (roughly mid-July through

September) among migrant bats species (see Johnson 2005). Some studies have shown apparent

low risk from turbines to resident bat populations (Johnson et al. 2003) while others have shownthat mortality is not correlated with AnaBat call rates (Nicholson 2002, 2003). The post-

construction mortality data collected at existing regional projects appears to be the best available

predictor of mortality levels and species composition for proposed wind projects. Some studiesof wind projects have recorded both AnaBat detections per night and bat mortality (Table 11).

The number of bat calls per night as determined from AnaBat detectors shows a rough

correlation with bat mortality but may be misleading because effort, timing of sampling, speciesrecorded, and detector settings (equipment and locations) varied among studies.

Table 11. Wind projects in the U.S. with both AnaBat sampling data and mortality data for batspecies.

Project Area

Study Period

Detector

nights

Bat activity

(#/detector/night)

Mortality

(bats/turbine/yr)

Reference

Mountaineer, WV Aug 1-Sep 14, 2004 33 38.3 38.0 Arnett 2005

Top of Iowa, IA Sep 4-Oct 9, 2003;

May 26-Sep 24, 2004

42 34.9 10.2 Koford et al.

2005

Foote Creek Rim, WY Jun 15-Sep 1, 2000-01 39 2.2 1.3 Gruver 2002

Buffalo Ridge, MN Jun 15-Sep 1, 2001 216 2.1 2.2 Johnson et al.2003

Buffalo Mountain, TN Apr 1-Sep 30, 2001-02 149 23.7 20.8 Fieldler 2004

The number of bats detected per night at the Cape Vincent met tower was highest in the summer.

Mortality studies of bats at wind projects in the U.S. have shown a peak in mortality in August

and September and generally lower mortality earlier in the summer (see Johnson 2005). Whilethe survey efforts varied among the different studies, the studies that included AnaBat surveys

and fatality surveys showed a general association between the timing of bat calls and timing of

mortality, with both peak call rates and peak mortality occurring during the fall (Table 11). Batactivity expressed as the average number of calls per detector-night recorded in the study area

(~0.65 to 28.7 bats per detector night) was not as high as the projects recording the highest batmortality (Table 11) and the highest call rate (~28.7 bats per detector-night) occurred in thesummer when bat mortality has typically been lower at other studies in the U.S. (Johnson 2005).

The nearest monitoring study of a wind project to the Cape Vincent site was the recent study at

the Maple Ridge project (Jain et al. 2007). Estimates of bat mortality at the Maple Ridge site

varied from 9.2 to 14.8 fatalities per MW (15.2 to 24.5 fatalities per turbine) (Jain et al. 2007).Pre-project summer bat activity recorded at the Maple Ridge site (20.6 calls per detector-hour;

Reynolds 2004) was higher than Cape Vincent (~3.2 calls per detector-hour for summer passive


56/59




sampling). This may indicate that bat mortality at Cape Vincent would be lower than Maple

Ridge, but summer bat mortality is generally lower at all wind projects studied including Maple

Ridge (Jain et al 2007, Johnson 2005). No AnaBat surveys were conducted in the fall at MapleRidge (Reynolds 2004) for comparison when bats are most at risk. Based on the AnaBat data

passage rates, it is not expected that bat mortality at Cape Vincent would be greater than thatreported at Maple Ridge. It is expected that bat mortality at the Cape Vincent project area will

be similar to the other studies in the U.S. with the peak of mortality likely occurring near lateAugust or early September. Spring and summer mortality levels for bats are expected to

bp sdeis app f-1 west survey reports cape vincent

Documents