Fig 1.
Locations of common ragweed occurrences throughout the eastern U.S.
Table 1.
Thirteen global climate models that were used in this study.
Fig 2.
Predicted current and future presence (occurrence) of common ragweed across the eastern U.S.
Left panel represents current predictions and the right panel represents the future distribution projections for RCP 4.5 and 8.5 and the 2050s and 2070s. The intensity of the colors represent agreement among the 13 global climate models.
Fig 3.
Model evaluation for common ragweed.
(A) Receiver operating characteristic (ROC) curve. Red line is for the training data (AUC = 0.780), blue line is for the test data (AUC = 0.765), and black line is for a random set of predictions (AUC = 0.5). (B) Omission and predicted area curve. Red line is the fraction of background that is predicted, blue line is the omission rate for the training samples, teal line is the omission rate for the test samples, and black line is the predicted omission.
Fig 4.
Response curves for the four climate variables used to predict common ragweed presence.
(A) January maximum temperature, (B) mean diurnal temperature range, (C) August precipitation, and (D) precipitation seasonality. Temperature variables in plots A and B have been multiplied by a factor of 10. For example, 5 degrees C is represented by “50”.
Table 2.
Future projected change in suitable area for common ragweed compared to the current distribution.