Table 2. Collision counts, results of removal and detection trials, and bias-adjusted fatality estimates for all buildings.
| Raw countsb | Bias trialsd | Bias-adjusted fatalitiese | ||||||
|---|---|---|---|---|---|---|---|---|
| Building Ida | Fatal | Non-fatal | # of speciesc | Removal | Detection | Detection prob. | Low | High |
| 17 | 254–264 | 51–51 | 44 | 43 | 5 | 0.59 (0.48–0.69) | 431 (370–525) | 448 (384–545) |
| 4 | 91–113 | 27–30 | 38 | 32 | 6 | 0.31 (0.18–0.45) | 297 (202–493) | 369 (251–613) |
| 1 (Stadium) | 155–159 | 70–70 | 42 | 27 | 23 | 0.70 (0.56–0.80) | 222 (192–274) | 228 (197–281) |
| 3 | 77–112 | 18–20 | 35 | 33 | 1 | 0.48 (0.36–0.61) | 158 (126–211) | 231 (184–307) |
| 8 | 4–8 | 0–1 | 4 | 5 | 2 | 0.04 (0.00–0.62) | 114 (6–4000) | 228 (12–8000) |
| 9 | 59–64 | 8–10 | 24 | 24 | 14 | 0.70 (0.58–0.81) | 83 (72–102) | 90 (78–111) |
| 19 | 29–34 | 4–4 | 17 | 14 | 3 | 0.43 (0.25–0.62) | 67 (46–115) | 79 (54–135) |
| 12 | 25–26 | 5–7 | 13 | 10 | 4 | 0.53 (0.33–0.72) | 47 (34–76) | 48 (36–79) |
| 20 | 23–28 | 9–10 | 15 | 9 | 2 | 0.51 (0.28–0.72) | 45 (32–83) | 54 (39–101) |
| 15 | 11–15 | 5–5 | 9 | 11 | 1 | 0.29 (0.14–0.50) | 37 (21–76) | 51 (29–104) |
| 13 | 6–8 | 4–4 | 9 | 8 | 4 | 0.24 (0.08–0.50) | 24 (12–72) | 32 (16–96) |
| 2 | 9–18 | 1–3 | 9 | 6 | 3 | 0.39 (0.18–0.64) | 22 (13–50) | 45 (27–101) |
| 6 | 14–45 | 4–4 | 10 | 14 | 4 | 0.64 (0.48–0.79) | 21 (17–29) | 70 (57–94) |
| 16 | 1–1 | 1–1 | 1 | 6 | 1 | 0.05 (0.00–0.61) | 20 (1–1000) | 20 (1–1000) |
| 21 | 5–7 | 2–2 | 5 | 8 | 9 | 0.45 (0.24–0.69) | 11 (7–20) | 15 (10–29) |
| 5 | 5–11 | 5–6 | 6 | 8 | 6 | 0.65 (0.44–0.81) | 7 (6–11) | 16 (13–24) |
| 10 | 4–4 | 3–3 | 6 | 5 | 6 | 0.56 (0.24–0.80) | 7 (4–16) | 7 (4–16) |
| 11 | 3–6 | 2–2 | 5 | 7 | 2 | 0.51 (0.24–0.74) | 5 (4–12) | 11 (8–24) |
| 14 | 1–6 | 1–1 | 2 | 3 | 3 | 0.18 (0.02–0.65) | 5 (1–58) | 33 (9–350) |
| 7 | 2–2 | 0–0 | 2 | 5 | 3 | 0.57 (0.29–0.78) | 3 (2–7) | 3 (2–7) |
| 18 | 2–2 | 0–0 | 2 | 8 | 3 | 0.61 (0.37–0.79) | 3 (2–5) | 3 (2–5) |
| Totals | 780–933 | 220–234 | 75 | 286 | 105 | - | 1629 (1170–7235) | 2081 (1413–12022) |
Collision counts, results of removal and detection trials, and bias-adjusted bird fatality estimates for 21 buildings, including U.S. Bank Stadium, monitored in downtown Minneapolis, Minnesota, 2017–2018. Table includes raw counts of fatal and non-fatal collisions; information about bias trials that were used to generate detection probability estimates accounting for both carcass removal and imperfect detection of window-killed bird carcasses; and bias-adjusted fatality estimates based on application of detection probability estimates to raw fatal collision counts. Buildings are ranked in descending order based on the low bias-adjusted fatality estimate (parentheses indicate 95% confidence intervals).
aUnique numeric code for each building used for purposes of current study.
bRaw counts for fatal and non-fatal collisions at each building; low and high values are counts that respectively exclude and include birds potentially resulting from predation events (for fatal collisions) and collisions with skyways between buildings (for fatal and non-fatal collisions).
cNumber of species observed as collision casualties across the entire study, including both fatal and non-fatal collisions.
dNumber of carcass removal trials conducted to quantify animal scavenger and human removal of carcasses, number of detection trials conducted to quantify surveyor detection probability for carcasses present in search area (excludes detection trials where trial carcasses were removed before surveyors had a chance to encounter them), and estimated probability of detecting a window-killed carcass that falls in the survey area (detection probability accounts for both removal and detection probability).
eBias-adjusted fatality estimates based on application of detection probability estimates to raw fatal collision counts; low and high adjusted estimates were generated using the low and high fatal collision counts.