Analysis of WSR-88D Data to Assess Nocturnal Bird Migration over the Lompoc Wind Energy Project in California

Transcription

1 3. RESULTS AND DISCUSSION 3.1 YEAR-TO-YEAR PATTERN OF MIGRATION The year-to-year pattern of nightly density of migratory movements derived from Level III base reflectivity files from the WSR-88D at Vandenberg AFB, CA for spring and fall can be found in Figures 4 and 5, respectively and in Appendix B; Tables B-1 and B-2, respectively. As expected, the density of migration measured by the maximum dbz method and the sample area (147 pixels) method differed. The former sampled higher altitudes while the latter sampled low altitudes where fewer birds migrated. Quite frequently in spring migration was occurring at altitudes above the altitudes sampled by the radar beam over the project area. Therefore it is not surprising that the density of migration based on maximum dbz of the reflectivity from migrating birds within the radar surveillance area was greater than that detected within the confines of the sample area of 147 pixels 3-36 km away from the radar (contrast the upper and lower graphs in Figure 4). The greatest density recorded in spring was 184 birds km -3 based on the maximum dbz method, but the greatest density recorded in spring within the sample area was only 86. It is also evident that the seasonal patterns of migration show year-to-year variation. More high-density movements occurred during the spring of 7 than during the spring of 6. Only one movement of 184 birds km -3 was recorded in 6 on 28 April, but six were detected in 7 between April and 7 May. Based on only two years of data, it appears that little nocturnal migration occurs in the spring before the middle of April and after the middle of May (Figure 4). In the fall the patterns of migration were somewhat similar between 6 and 7. In both years more migration occurred between 15 August and 3 September than between 1 October and 15 November. In the fall of 6 the densest flights were recorded on September (184 birds km -3 ), and in 7 the densest flights of the fall occurred on 21 and September (463 birds km -3 ). In the fall the mean density of birds in the sample area of 147 pixels 3-36 km away from the radar was with a few exceptions closer to that estimated by the maximum dbz method (compare the upper and lower graphs of Figures 4 and 5). The reason for this difference between spring and fall is likely related to wind patterns. In the fall the winds are typically from the W, NW, and N at lower altitudes. In spring the winds are essentially the same at lower altitudes, but at higher altitudes (15 m and higher) the winds blow from the SE and S. Thus migrants that fly higher in the spring do so with favorable winds and avoid the adverse winds at lower altitudes. Flying at higher altitudes in the fall is not necessary because winds at lower altitudes are frequently favorable. 3.2 NIGHT-TO-NIGHT PATTERN OF MIGRATION Data on the night-to-night pattern of nocturnal migration over the project area for spring (6-7) and fall (6-7) can be found in Appendix B, Tables B1 and B2, respectively. Nocturnal migration during the spring and fall shows considerable night-to-night variability. In the spring (Figure 4), migration begins to build in mid-april, peaks near the end of April and the beginning of May, and then declines after the first week of May. Dates with a mean density of 1 birds km -3 or greater during the two year study occurred from 19 April and 8 May (Figure 4, upper graph), and although sizable flights can occur anytime from the middle of April through the middle of May, the peak of migration through the project area is in late April and early May. Nocturnal migration during the fall also varies from night-to-night (Figure 5). A pulse of fall migration builds in late August and early September, and another pulse of greater magnitude occurs in late September. From the beginning of October the density of migration declines, and by November very little migratory movement takes place. In the spring 12 flights exceeded 1 birds km -3 or greater, and in the fall, 1 flights exceeded 1 birds km -3 or greater, but the highest densities of migration occurred in the fall. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 6

2 16 1 Spring 6 Max dbz Spring 7 Max dbz Mar 17-Mar 19-Mar 21-Mar 23-Mar 25-Mar 27-Mar 29-Mar 31-Mar 2-Apr 4-Apr 6-Apr 8-Apr 1-Apr 12-Apr 14-Apr 16-Apr 18-Apr -Apr 22-Apr 24-Apr 26-Apr 28-Apr 3-Apr 2-May 4-May 6-May 8-May 1-May 12-May 14-May 16-May 18-May -May 22-May 24-May 26-May 28-May 3-May Birds per Cubic KM 1 Date Spring 6 Sample Area Spring 7 Sample Area 1 15-Mar 17-Mar 19-Mar 21-Mar 23-Mar 25-Mar 27-Mar 29-Mar 31-Mar 2-Apr 4-Apr 6-Apr 8-Apr 1-Apr 12-Apr 14-Apr 16-Apr 18-Apr -Apr 22-Apr 24-Apr 26-Apr 28-Apr 3-Apr 2-May 4-May 6-May 8-May 1-May 12-May 14-May 16-May 18-May -May 22-May 24-May 26-May 28-May 3-May Mean Birds per Cubic KM Date Figure 4. Year-to-year and night-to-night comparison of birds km -3 for the spring seasons of 6 and 7. Upper figure: bird density (birds km -3 ) based on maximum dbz of bird migration reflectivity in radar surveillance area. Lower figure is mean bird density (birds km - 3 ) within the 147 pixel sample area over the wind project site. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 7

3 Fall 6 Max dbz Fall 7 Max dbz Aug 17-Aug 19-Aug 21-Aug 23-Aug 25-Aug 27-Aug 29-Aug 31-Aug 2-Sep 4-Sep 6-Sep 8-Sep 1-Sep 12-Sep 14-Sep 16-Sep 18-Sep -Sep 22-Sep 24-Sep 26-Sep 28-Sep 3-Sep 2-Oct 4-Oct 6-Oct 8-Oct 1-Oct 12-Oct 14-Oct 16-Oct 18-Oct -Oct 22-Oct 24-Oct 26-Oct 28-Oct 3-Oct 1-Nov 3-Nov 5-Nov 7-Nov 9-Nov 11-Nov 13-Nov 15-Nov Birds per Cubic KM Date Fall 6 Sample Area Fall 7 Sample Area 15-Aug 17-Aug 19-Aug 21-Aug 23-Aug 25-Aug 27-Aug 29-Aug 31-Aug 2-Sep 4-Sep 6-Sep 8-Sep 1-Sep 12-Sep 14-Sep 16-Sep 18-Sep -Sep 22-Sep 24-Sep 26-Sep 28-Sep 3-Sep 2-Oct 4-Oct 6-Oct 8-Oct 1-Oct 12-Oct 14-Oct 16-Oct 18-Oct -Oct 22-Oct 24-Oct 26-Oct 28-Oct 3-Oct 1-Nov 3-Nov 5-Nov 7-Nov 9-Nov 11-Nov 13-Nov 15-Nov Mean Birds per Cubic KM Date Figure 5. Year-to-year and night-to-night comparison of birds km -3 for the fall seasons of 6 and 7. Upper figure: bird density (birds km -3 ) based on maximum dbz of bird migration reflectivity in radar surveillance area. Lower figure is mean bird density (birds km - 3 ) within the 147 pixel sample area over the wind project site. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 8

4 The density of migration appears to be lower than that observed with WSR-88D in other areas of the United States. In a WSR-88D study of bird migration over Eglin Air Force Base in northwestern Florida during the fall of 5, Gauthreaux et al. (7) recorded a peak of migration on 14 October of 1148 birds km -3. The mean density for the entire fall season was 142 birds km -3 (standard deviation 197 birds km -3, and coefficient of variation = 1.39). In a study of bird migration over Grand Forks Air Force Base in North Dakota for the years Gauthreaux et al. (5) found that the temporal pattern of spring migration (-3) peaked usually between the first few days of May and the last week in May, and little migration occurred after the first week in June. Peak densities were 5-6 birds km -3. In the fall the seasonal temporal pattern of bird migration over Grand Forks, North Dakota, was relatively similar from year to year but the numbers of birds varied annually. Movements of + birds km -3 began after the second week of August, peak flights above birds km -3 generally began near the middle of September, and most of the dense flights (+ birds km -3 ) were completed by the first few days of October. By late October only smaller flights occurred until about the first week in November. After the first week in November migration was essentially over for the fall. 3.3 HOUR-TO-HOUR PATTERN OF MIGRATION The data showing the hour-to-hour pattern of migration over the project area during the spring (6-7) and fall (6-7) are in Appendix C; Tables C-1 and C-2, respectively. In spring, migration typically started 3-45 minutes after sunset, peaked on most evenings between 5: 9: UTC (9: PM 1: AM Pacific Standard Time [PST]), and declined until sunrise (Figure 6). In the fall (Figure 7) the quantity of migration was greater than in the spring (see above), and the hour-to-hour pattern of percentage of peak hourly density during the evenings was shifted slightly earlier in the evening compared to that observed in spring (compare Figures 6 and 7). The peak of a nightly movement generally occurred from 3: 4: UTC (7: PM 8: PM PST). The hour-to-hour temporal patterns of migration found in this study are similar to those found in other studies (Newman 1956, Farnsworth et al 4, Black, and Gauthreaux et al. 5, 7). The hour-to-hour temporal patterns recorded by Newman (1956) were based on moon-watching, and Newman found that "flight densities tend to reach their nightly crest a bit earlier in autumn than in spring and the peak is not quite as sharply defined." In this study the timing of the peak densities in the fall appears to agree with Newman's findings. 3.4 DIRECTION OF MIGRATORY MOVEMENTS The directional tendencies of the migratory flights over the project area are given in Tables 2 and 3 and in Figure 8. In the spring the mean directions (μ) from which the movements originated were in 6 and in 7 and the flights were oriented toward the north- northwest (between 342 and 343. ). There was little variability in direction and all showed relatively strong directionality (length of the mean vector [r] and concentration Table 2, Figure 8). All yearly mean directions show low circular variance and are highly significant (p <.). In the fall the mean directions were from in 6 and in 7 and the flights were oriented toward the south between 152 and 153. The lengths of the mean vectors (r), a measure of tightness of direction, were comparable (Table 3) to those in spring (Table 2) suggesting that the directions of fall flights were in essentially the same direction. This is illustrated by comparing the circular diagrams in Figures 8. The fall flights are within 1 of the back azimuth direction of spring flights. Topographic features such as the coastline and coastal mountain ridges likely influence the directions of seasonal migrations, particularly those occurring at lower altitudes. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 9

5 1.% 9.% 8.% 7.% 6.% 5.%.% 3.%.% 1.%.% 17: 18: 19: : 21: 22: 23: : 1: 2: 3: 4: Mean Percent of Peak 5: Pacific Standard Time Figure 6. Hour-to-hour pattern of bird migration in the spring (6-7). The bars show mean percentage of peak mean density (birds km -3 ) and the lines indicate the standard error of the mean. Mean Percent of Peak 1.% 9.% 8.% 7.% 6.% 5.%.% 3.%.% 1.%.% 17: 18: 19: : 21: 22: 23: : 1: 2: 3: 4: 5: Pacific Standard Time Figure 7. Hour-to-hour pattern of bird migration in the fall (6-7). The bars show mean percentage of peak mean density (birds km -3 ) and the lines indicate the standard error of the mean. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 1

6 Table 2. Statistical analysis of direction of nightly migrations over the Lompoc Wind Energy Project during the spring (6-7). Year Variable Spring 6 Spring 7 Number of Observations Mean Vector (µ) Length of Mean Vector (r) Circular Variance 1.1E-2.25 Circular Standard Deviation Standard Error of Mean Rayleigh Test (Z) Rayleigh Test (p) Rao's Spacing Test (U) Rao's Spacing Test (p) <.1 <.1 Table 3. Statistical analysis of direction of nightly migrations over the Lompoc Wind Energy Project during the fall (6-7). Year Variable Fall 6 Fall 7 Number of Observations 59 8 Mean Vector (µ) Length of Mean Vector (r) Circular Variance Circular Standard Deviation Standard Error of Mean Rayleigh Test (Z) Rayleigh Test (p) Rao's Spacing Test (U) Rao's Spacing Test (p) <.1 <.1 This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 11

7 Spring 6 Spring Fall 6 Fall Figure 8. Circular diagrams showing the direction of spring and fall migration for the years 6-7. The statistical details are in Tables 4 and 5. The dark line is the mean angle and the arc at the end is the 95% confidence limits of the mean. This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 12

8 3.5 MIGRATION, WEATHER CONDITIONS, AND COLLISIONS In an attempt to document the number of nights during spring and fall when the altitude of migration might be lowered because of low ceilings and visibility would be obscured by fog or precipitation, we recorded the number of nights with overcast and ceiling below 1 ft. during spring (15 March- 31 May 6-7) and fall (15 August - 15 November, 6-7) and any of the following weather conditions during the nighttime hours: fog, mist, drizzle, light rain, moderate rain, heavy rain. We also recorded the density of migration on these nights (Table 6, Appendix D; Tables 1-D and D-2). During the spring of 6, 21 of 78 nights had conditions that would cause birds to fly lower -- sometimes with reduced visibility. Of these nights only four had migration densities of 25 birds km -3 or greater. In the spring of 7, 35 of 78 nights had low ceiling conditions and of these nights only five had bird movements of 25 birds km -3 or greater (Table 6). During the fall of 6, 45 of 93 nights had conditions that might cause birds to migrate at low altitudes in poor weather and of these nights 14 had bird movements of 25 birds km -3 or greater (Table 6). In the fall of 7, 41 of 93 nights examined had conditions that might cause birds to migrate at low altitudes in poor weather and of these nights four had bird movements of 25 birds km -3 or greater There were more nights in the fall than in spring with weather conditions that could cause birds to fly at low altitudes and sometimes in poor visibility, but on the vast majority of these nights the amount of migration was greatly reduced or there was no migration taking place. Table 4. Number of nights in spring (6-7) and in fall (6-7) when the density of migration was 25 birds km -3 or greater and the ceiling was 1 ft or lower, the sky was overcast, and one or more of the following weather conditions existed during the nighttime hours: fog, mist, drizzle, light rain, moderate rain, heavy rain. Night with low ceilings Season and weather Nights with 25 birds km -3 or more Spring Fall DISCUSSION Based on the analysis of two years of archived radar data from the WSR-88D radar at Vandenberg AFB, the amount of bird migration occurring within 1 nautical miles of the radar is much lower than that recorded by WSR-88D radars in other areas of the United States. Based on a comparison of peak migration densities at 7+ WSR-88D radar sites throughout the United States (Gauthreaux et al. 3), the peak densities of bird migration are lower in the West and along the West Coast than elsewhere in the United States (e.g., the Great Plains and the East) and the densities recorded at the Vandenberg site are typical for those recorded along the West Coast. Within the surveillance area of the Vandenberg radar (Figure 3), nocturnal migration appeared to be heavier east of the project area than over the project area. Mean peak migration densities were typically lower in the sample area (over the project area) than within the surveillance area of the Vandenberg radar (Figure 3). This is not surprising because the altitudes sampled by the radar beam over the project area were lower (Table 1) than the altitudes at which migrating birds typically fly. As the radar beam increased in altitude higher densities of migrants were recorded. One must use caution in applying the sample area densities to birds passing through the rotor swept zone (RSZ). The radar beam sampled approximately 5 % of the RSZ, but 5% of the radar beam was above the RSZ. It is impossible to tell exactly where the birds This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 13

9 are within the radar beam over the sample area, but based on the fact that migration density increased with altitude, it is likely that a major portion of the birds recorded in the sample area were above the RSZ. Based on the analysis presented in this report, the Lompoc Wind Energy Project should have minimal impact on migrating birds. 4. LITERATURE CITED Able, K. P A radar study of the altitude of nocturnal passerine migration. Bird-Banding 41: Avery, M., P. F. Springer, and J. F. Cassel Weather influences on nocturnal bird mortality at a North Dakota tower. The Wilson Bulletin 89: Bellrose, F. C The distribution of nocturnal migrants in the air space. Auk 88: Berthold, P Bird Migration. A general survey. Oxford University Press, Oxford. Black, J. E.. Application of weather radar to monitoring numbers of birds in migration. Online Crawford, R. L Weather, migration and autumn bird kills at a north Florida TV tower. Te Wilson Bulletin 93: Crum, T.D., and R.L. Alberty The WSR-88D and the WSR-88D operational support facility. Bulletin of the American Meteorological Society 74: Bruderer, B, and A. Boldt. 1. Flight characteristics of birds: 1. radar measurements of speeds. Ibis 143: Crum, T.D., R.L. Alberty, and D.W. Burgess Recording, archiving, and using WSR-88D data. Bulletin of the American Meteorological Society 74: Diehl, R.H., and R.P. Larkin. 5. Introduction to the WSR-88D (NEXRAD) for ornithological research. Pp in Bird Conservation Implementation and Integration in the Americas: Proceedings of the Third International Partners in Flight Conference 2 March -24, Asilomar, California; Volume 2, C.J. Ralph and T.D. Rich, eds. Gen. Tech. Rep. PSW-GTR-191. Albany, CA: Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture. Dutcher, W Bird notes from Long Island, N.Y. Auk 1:31-35 Farnsworth, A., S. A. Gauthreaux, Jr., and D. E. Van Blaricom. 4. A comparison of nocturnal call counts of migrating birds and reflectivity measurements on Doppler radar. Journal of Avian Biology 35: Gauthreaux, S.A., Jr., and C.G. Belser Displays of bird movements on the WSR-88D: Patterns and quantification. Weather and Forecasting 13: Gauthreaux, S.A., Jr., and C.G. Belser Reply to displays of bird movements on the WSR-88D: Patterns and quantification. Weather and Forecasting 14: Gauthreaux, S.A., Jr., and C.G. Belser. 3. Bird movements on Doppler weather surveillance radar. Birding 35: Gauthreaux, S.A., Jr., C.G. Belser, and A. Farnsworth.. How to use Doppler weather-surveillance radar to study hawk migration. Pp in Hawkwatching in the Americas, K.L. Bildstein and D. Klem, Jr., eds. North Wales, Pennsylvania: Hawk Migration Associateion of North America. Gauthreaux, S.A., Jr., C.G. Belser, and D.Van Blaricom. 3. Using a network of WSR 88-D weather surveillance radars to define patterns of bird migration at large spatial scales. Pp in Avian Migration, P. Berthold, E. Gwinner, and E. Sonnenschein, eds. Germany: Springer-Verlag. Gauthreaux, S. A., Jr., C. G. Belser, J. W. Livingston, and D. Van Blaricom. 7. The identification of military installations as important migratory bird stopover sites and the development of bird migration forecast models: A radar ornithology approach. Annual Report of Research Accomplishments, Strategic Environmental Research and Development Program (SERDP), U. S. Department of Defense. January 7. Gauthreaux, S. A., Jr., M. P. Guilfoyle, R. A. Fischer, and G. W. Fleming. 5. Seasonal bird surveys and the development of bird migration forecast models for Grand Forks Air Force Base, ND. Final Report. Prepared for the U.S. Air Force, Grand Forks Air Force Base, North Dakota. March 5. Klazura, G.E. and D.A. Imy A description of the initial set of analysis products available from the NEXRAD WSR-88D System. Bulletin American Meteorological Society 73: Lack, D The height of bird migration. Brit. Birds 53: 5-1. Mabee, T. J., and B. A. Cooper, J. H. Plissner, D. P. Young. 6. Nocturnal bird migration over an Appalachian ridge at a proposed wind power project. Wildlife Society Bulletin 34: This Geo-Marine, Inc. document is privileged, company proprietary data, and is intended only for the addressees use. 14