MEASUREMENTS OF THE SUSCEPTIBILITY OF PIONEER TREES TO UPROOTING USING PULL TESTS

Pioneer woody riparian trees may be uprooted by river flows if they are subjected to drag forces greater than their anchoring ability. To simulate seedlings’ ability to withstand uprooting, we measured pullout force for cottonwood (Populus), willow (Salix), and tamarisk (Tamarix) seedlings (n = 350; 1-5 yrs. old) at three sites characterized by different hydrology and substrates, including an unregulated gravel-bed (Bitterroot River, MT), an unregulated sand-bed (Santa Maria River, AZ), and an impounded sand-bed river (Bill Williams River, MT). To assess the decrease in anchoring ability caused by scour, we excavated the substrate around plants to depths of 10 – 40 cm for a subset of seedlings (n = 100) before uprooting. We conducted statistical analyses to compare the uprooting force for each species and site. For unscoured plants, ANCOVA accounting for seedling size (pullout force vs. frontal area) indicates plants are most easily uprooted at the impounded sand-bed site (p < 0.001). We compared measured pullout force to a range of drag forces. For unscoured plants, pullout force is >> drag force for all but the smallest seedlings (1-2 yrs. old), indicating seedlings are unlikely to be uprooted in the absence of scour. To assess the influence of the scour treatments on the pullout force vs. frontal area relationship, we compared the scoured plants to the unscoured plants for each site. For the sand-bed sites, the 20 cm scour treatment resulted in a decrease in slope of the pullout force vs. frontal area relationship for tamarisk (interaction term; p < 0.01), whereas the relationship for scoured cottonwood could not be differentiated from unscoured seedlings (interaction term and intercept; p > 0.1). Drag forces exceeded pullout forces for 20 cm scoured tamarisk seedlings, but did not exceed pullout forces of scoured cottonwood with the exception of very small seedlings. The implications of this study are that: 1) elevated static water table regimes may cause a decrease in uprooting resistance; 2) uprooting without significant scour is unlikely; 3) in floods where moderate scour occurs, native cottonwood may be more resilient to uprooting than tamarisk.