Table 1.
Comparison of experiments testing songbirds for disorientation in monofrequent radio-frequency (RF) magnetic fields at the local Larmor frequency of the free electron
| Study | [1], [2] | [3] | [4] |
|---|---|---|---|
| Location | Frankfurt, Germany | Courish Spit, Russia | Oldenburg, Germany |
| Bird species | Erithacus rubecula | Sylvia borin | Erithacus rubecula |
| Local magnetic induction (nT) | Between 46,000 and 47,400 | 50,100 | 48,600 ± 240 |
| Local Larmor frequency (MHz) | Between 1.289 and 1.328 | 1.404 | 1.362 ± 0.007 |
| RF test condition | [1] 485 nT@1.315 MHz [2] 15 nT@1.315 MHz |
190 nT@1.403 MHz | (a) 400 nT@1.363 MHz (b) 48 nT@1.363 MHz |
| Confound control (sham-RF) | None | None | (a) 400 nT@ 50 Hz (b) RF power fed into dummy load |
| Birds per condition | [1] 12 (spring), 16 (autumn) [2] 12 |
8 | (a) 19, 20 (b) 31 |
| Trails per bird in each condition | 3 | 3 | 10 |
| Rayleigh test (second order, RF cond) | [1] p > 0.05 (n.s.) [2] p > 0.05 (n.s.) |
p = 0.83 (n.s.) | (a) p < 0.01 (**) (b) p = 0.138 (n.s.) |
| MWW: RF vs ctrl | [1] p < 0.001 (***) [2] p < 0.001 (***) |
p = 0.025 (*) | (a) p = 0.41 (n.s) (b) p = 0.77 (n.s) |
| RF generator | Stanford Research DS340 | Not specified | Rigol DG1022 |
| Frequency standard | None | None | None |
| RF power amplifier | Amplifier Research 25 W 1–1000 MHz | Not specified | TOMCO, 50 W CW 0.1–20 MHz |
| Spectrum analysis | HP 89410A (DC—10 MHz) | Digital oscilloscope FFT | Rohde and Schwarz FSV-3 (10 Hz–3.6 GHz) |
| Background RFlevel | Not reported | B rms < 0.5 nT | Figure 3 in [4], E rf and B rf |
| Emlen funnel | Plastic (PVC) | Cardboard | Nonmagnetic metal (aluminum) |
| Trial duration | 75 min | 40 min | 60 min |
| Light conditions | Monofrequent 565 nm 2.1 mW/m2 | Blurred night sky (outdoor) | White (incandescent bulb) 2.1 mW/m2 |
Study: [1] Thalau et al. (2005); [2] Ritz et al. (2009); [3] Kavokin et al. (2014); [4] Schwarze et al. (2016)
Larmor frequency refers to the “free” electron with a g-factor of 2.0023
Rayleigh test: test against null hypothesis that directions were randomly drawn from a uniform distribution over the unit circle; low p value suggests that null hypothesis is false, i.e., that mean direction is significant
MWW (Mardia–Watson–Wheeler test): non-parametric test against null hypothesis that both experimental and control distribution are drawn from the same distribution, low p value suggests that null hypothesis is false, i.e., that the distributions are not the same