Online Resource 1: (A) Distribution of Average Daily Wind Directions (In Radians). Box

Online Resource 1: (a) Distribution of average daily wind directions (in radians). Box and whisker plot showing the distribution of daily measures of wind direction combined for years 2001-2004 and 2006-2008. The horizontal dashed line represents the average wind direction for the whole period.

A description

Online Resource 2. Partial relationship between patch area and detectability for (a) rock and (b) tree phorophyte for the best-supported model.

Macintosh HD Users GalloRojo detectability pdf

Online Resource 3: Theoretical distributions of the parameter δij. A random spatial arrangement (A) results in a relatively uniform distribution (B). If patches are aligned almost perfectly in the same angle of wind direction (C), it results in a bimodal distribution highly skewed to the limits of the distribution (D).

Macintosh HD Users GalloRojo Desktop Working folder UF LabStuff Lepanthes MS theoretical distributions pdf


Online Resource 4: (a) Spatial representation of relative patch areas in the L. rupestris metapopulation. (b) Distribution of patch elevations (z coordinate) for both phorophytes. Panes (c) and (d) show the distribution of patch sizes for rock and tree phorophytes respectively.XGH Users migdaliselcolon berlingeri Desktop work UF LabStuff Lepanthes colext area elevation pdf

Online Resource 5. Distribution of the δ parameter in the Lepanthes system based on the average wind direction of all sampling periods (2.15 rad). It describes the deviation between wind direction and the angle between patches with respect to the horizontal axis. Bimodal distributions represent conditions where asymmetric and symmetric measures of connectivity should deviate most (Online Resource 3). This parameter is incorporated in the connectivity measure that accounts for potential asymmetries in effective distance.

SCIENCEWORK Working folder UF LabStuff Lepanthes MS distribution fig pdf

Online Resource 6: Parameter estimates and standard errors (SE) for the dynamic occupancy model that resulted in the best fit; ψ(.), ϒ(Sasym+A+Ph), ε(A), P(A+Ph).

Colonization / Extinction / Detection
Estimate / SE / Estimate / SE / Estimate / SE
INTERCEPT / -4.540 / 0.171 / -3.181 / 0.137 / 2.332 / 0.084
Sasym / 0.360 / 0.092 / - / - / - / -
A / 0.513 / 0.073 / -0.808 / 0.240 / 0.290 / 0.074
Ph (TREES) / 0.717 / 0.256 / - / - / -0.428 / 0.134

Online Resource 7: Parameter estimates and standard errors (SE) for the dynamic occupancy model most similar to the best-ranked model, but with the patch connectivity measure that included a symmetric dispersal kernel (Ssym) as a site covariate, ψ(.), ϒ(Ssym+A+Ph), ε(A), P(A+Ph).

Colonization / Extinction / Detection
Estimate / SE / Estimate / SE / Estimate / SE
INTERCEPT / -4.531 / 0.171 / -3.187 / 0.137 / 2.328 / 0.084
Sasym / 0.273 / 0.106 / - / - / - / -
A / 0.501 / 0.074 / -0.795 / 0.240 / 0.294 / 0.074
Ph (TREES) / 0.714 / 0.259 / - / - / -0.431 / 0.134

Online Resource 8: Comparison of model fit for dynamic occupancy models using symmetric (Ssym), asymmetric connectivity (Sasym), phorophyte type (Ph) and patch area (A) as site covariates. This table compares all models fitted for each hypothesis.

Hypothesis / Model / Parameters / AIC / ΔAIC / AICweight
Asymmetric: Target / Ψ(.), γ(Sasym+A+Ph), ε(A), p(A+Ph) / 10 / 4321.65 / 0 / 0.43
Asymmetric: Target / Ψ(.), γ(Sasym+A+Ph), ε(A+Ph), p(A+Ph) / 11 / 4322.90 / 1.25 / 0.23
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym+A+Ph), ε(Sasym +A), p(A+Ph) / 11 / 4323.46 / 1.81 / 0.17
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym+A+Ph), ε(Sasym +A+Ph), p(A+Ph) / 12 / 4324.76 / 3.11 / 0.09
Symmetric: Target / Ψ(.), γ(Ssym+A+Ph), ε(A), p(A+Ph) / 10 / 4327.69 / 6.04 / 0.02
Asymmetric: Target / Ψ(.), γ(Sasym+A), ε(A+Ph), p(A+Ph) / 10 / 4328.66 / 7.01 / 0.01
Symmetric: Target / Ψ(.), γ(Ssym+A+Ph), ε(A+Ph), p(A+Ph) / 11 / 4328.97 / 7.32 / 0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym+A+Ph), ε(Ssym +A), p(A+Ph) / 11 / 4329.69 / 8.04 / <0.01
Asymmetric: Target / Ψ(.), γ(Sasym+A+Ph), ε(A+Ph), p(A) / 10 / 4329.87 / 8.22 / <0.01
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym+A), ε(Sasym +A+Ph), p(A+Ph) / 11 / 4330.50 / 8.85 / <0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym+A+Ph), ε(Ssym +A+Ph), p(A+Ph) / 12 / 4330.96 / 9.32 / <0.01
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym+A+Ph), ε(Sasym +A+Ph), p(A) / 11 / 4331.77 / 10.12 / <0.01
No connectivity / Ψ(.), γ(A+Ph), ε(A+Ph), p(A+Ph) / 10 / 4333.04 / 11.39 / <0.01
Symmetric: Target / Ψ(.), γ(Ssym+A), ε(A+Ph), p(A+Ph) / 10 / 4334.46 / 12.82 / <0.01
Symmetric: Target / Ψ(.), γ(Ssym+A+Ph), ε(A+Ph), p(A) / 10 / 4335.98 / 14.33 / <0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym+A), ε(Ssym +A+Ph), p(A+Ph) / 11 / 4336.46 / 14.82 / <0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym+A), ε(Ssym +A+Ph), p(A) / 11 / 4337.97 / 16.32 / <0.01
Asymmetric: Target / Ψ(.), γ(Sasym+A+Ph), ε(A+Ph), p(Ph) / 10 / 4341.91 / 20.27 / <0.01
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym+A+Ph), ε(Sasym +A+Ph), p(Ph) / 11 / 4343.80 / 22.16 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym+Ph), ε(A), p(A+Ph) / 9 / 4344.45 / 22.80 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym+Ph), ε(A+Ph), p(A+Ph) / 10 / 4345.77 / 24.13 / <0.01
Asymmetric: Rescue / Ψ(.), γ(Sasym+Ph), ε(Sasym +A+Ph), p(A+Ph) / 11 / 4347.69 / 26.04 / <0.01
Symmetric: Target / Ψ(.), γ(Ssym+A+Ph), ε(A+Ph), p(Ph) / 10 / 4348.44 / 26.80 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym+Ph), ε(A), p(A+Ph) / 9 / 4349.13 / 27.48 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym), ε(A+Ph), p(A+Ph) / 9 / 4350.35 / 28.70 / <0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym+A+Ph), ε(Ssym +A+Ph), p(Ph) / 11 / 4350.43 / 28.78 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym+Ph), ε(A+Ph), p(A+Ph) / 10 / 4350.47 / 28.83 / <0.01
Symmetric: Rescue / Ψ(.), γ(Ssym+Ph), ε(Ssym +A), p(A+Ph) / 10 / 4351.12 / 29.48 / <0.01
Asymmetric: Rescue / Ψ(.), γ(Sasym), ε(Sasym +A+Ph), p(A+Ph) / 10 / 4352.24 / 30.59 / <0.01
Symmetric: Rescue / Ψ(.), γ(Sasym + Ph), ε(Sasym +A+Ph), p(A+Ph) / 11 / 4352.46 / 30.82 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym + Ph), ε(A+Ph), p(A) / 9 / 4352.82 / 31.17 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym), ε(A+Ph), p(A+Ph) / 9 / 4354.63 / 32.98 / <0.01
Asymmetric: Rescue / Ψ(.), γ(Sasym + Ph), ε(Sasym +A+Ph), p(A) / 10 / 4354.77 / 33.12 / <0.01
Asymmetric: Target / Ψ(.), γ(Sasym +A+ Ph), ε(A+Ph), p(.) / 9 / 4356.10 / 34.46 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym + Ph), ε(A+Ph), p(A) / 9 / 4357.56 / 35.91 / <0.01
Symmetric: Rescue / Ψ(.), γ(Ssym), ε(Ssym +A+Ph), p(A+Ph) / 10 / 4357.68 / 36.03 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym + Ph), ε(A+Ph), p(Ph) / 9 / 4357.84 / 36.19 / <0.01
Asymmetric: Target-Rescue / Ψ(.), γ(Sasym + A + Ph), ε(Sasym +A+Ph), p(.) / 10 / 4358.10 / 36.45 / <0.01
Symmetric: Rescue / Ψ(.), γ(Ssym + Ph), ε(Ssym +A+Ph), p(A) / 10 / 4359.55 / 37.90 / <0.01
Asymmetric: Rescue / Ψ(.), γ(Sasym + Ph), ε(Sasym +A+Ph), p(Ph) / 10 / 4359.84 / 38.19 / <0.01
Symmetric: Target / Ψ(.), γ(Ssym +A+ Ph), ε(A+Ph), p(.) / 9 / 4362.77 / 41.13 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym +Ph), ε(A+Ph), p(Ph) / 9 / 4362.93 / 41.29 / <0.01
Symmetric: Target-Rescue / Ψ(.), γ(Ssym +A+Ph), ε(Ssym +A+Ph), p(.) / 10 / 4364.64 / 42.99 / <0.01
Symmetric: Rescue / Ψ(.), γ(Ssym +Ph), ε(Ssym +A+Ph), p(Ph) / 10 / 4365.72 / 44.08 / <0.01
Asymmetric: Area-Isolation / Ψ(.), γ(Sasym +Ph), ε(A+Ph), p(.) / 8 / 4370.86 / 49.21 / <0.01
Asymmetric: Rescue / Ψ(.), γ(Sasym +Ph), ε(Sasym +A+Ph), p(.) / 9 / 4372.84 / 51.20 / <0.01
Symmetric: Area-Isolation / Ψ(.), γ(Ssym +Ph), ε(A+Ph), p(.) / 8 / 4375.89 / 54.25 / <0.01
Symmetric: Rescue / Ψ(.), γ(Ssym +Ph), ε(Ssym +A+Ph), p(.) / 9 / 4378.69 / 57.04 / <0.01
NULL / Ψ(.), γ(.), ε(.), p(.) / 4 / 4418.17 / 96.52 / <0.01

Online Resource 9. Predicted partial relationships between colonization, extinction and patch area for both phorophytes for the best-supported model.

SCIENCEWORK Working folder UF LabStuff Lepanthes colext Supp figs area pdf